Menu
▾
▴
cdao-discuss — Discussion of design, implementation, evaluation, and application of CDAO
You can subscribe to this list here.
| 2009 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
(5) |
Oct
(11) |
Nov
(12) |
Dec
(14) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2010 |
Jan
(10) |
Feb
(33) |
Mar
(10) |
Apr
(2) |
May
(3) |
Jun
(2) |
Jul
(3) |
Aug
(13) |
Sep
(3) |
Oct
(12) |
Nov
(5) |
Dec
(2) |
| 2011 |
Jan
|
Feb
|
Mar
(3) |
Apr
(11) |
May
|
Jun
|
Jul
(4) |
Aug
(8) |
Sep
(8) |
Oct
|
Nov
(24) |
Dec
|
| 2012 |
Jan
|
Feb
(8) |
Mar
(2) |
Apr
|
May
|
Jun
(3) |
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| 2013 |
Jan
|
Feb
(2) |
Mar
(3) |
Apr
(2) |
May
(1) |
Jun
(1) |
Jul
(4) |
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: Enrico P. <epo...@cs...> - 2011-09-19 14:57:08
|
I'd be happy and interested in continuing supporting the development of CDAO. I agree with Rutger the there are problems that need to be addressed asap. Count me in. Enrico -- Dept. Computer Science, New Mexico State University MSC CS, Box 30001, Las Cruces, NM 88003 Voice: 575-646-6239 Fax: 575-646-1002 From: Christian Zmasek <cmz...@ya...> Reply-To: Christian Zmasek <cmz...@ya...> Date: Thu, 15 Sep 2011 10:55:26 -0700 (PDT) To: Rutger Vos <R....@re...>, CDAO list <cda...@li...>, "NeXML-discuss (list)" <nex...@li...>, Hilmar Lapp <hl...@ne...>, MIAPA <mia...@go...> Subject: Re: [CDAO-discuss] [Nexml-discuss] CDAO status and future Hi, Rutger: To be honest, I have not been following CDAO, so I am not entirely clear what it encompasses. In any case, if you think my expertise on molecular evolution and comparative genomics could be useful in revitalizing CDAO, I'd be happy to contribute. Christian > > From: Rutger Vos <R....@re...> > To: CDAO list <cda...@li...>; NeXML-discuss (list) > <nex...@li...>; Hilmar Lapp <hl...@ne...>; MIAPA > <mia...@go...> > Sent: Wednesday, September 14, 2011 3:06 AM > Subject: Re: [Nexml-discuss] CDAO status and future > > Hi all, > > it seems to me that it is critically important for a number of > projects we are all involved in (HIP, MIAPA, TreeBASE, etc.) that > there is an ontology for evolutionary informatics whose developers > respond to user needs. I dearly hope someone will step forward to push > further CDAO development. > > If not, I guess it might mean we will have to resort to a combination > of EDAM, MAO, and other ontologies - although those might not respond > as much to user requests as we need at this point in the game. > > Rutger > > On Tue, Sep 6, 2011 at 11:41 AM, Rutger Vos <R....@re...> wrote: >> > Hi all, >> > >> > I hate to be blunt, but I'm worried about CDAO: it's broken beyond the >> > point of usability and it doesn't seem to be maintained. I've now >> > tried to do some serious work with it, querying RDF graphs to fish out >> > the taxa, character state matrices and trees and it is simply not >> > working correctly. Some examples: >> > >> > * characters (i.e. columns) do have an ordering that needs to be >> > maintained, especially for molecular data, this should either be done >> > using column numbers or using an rdf list in the character >> > specification. (Arguably, orderings in other things - lists of taxa, >> > trees - would be useful too.) >> > >> > * it needs to be discoverable what data type a matrix is. It is no >> > good having to go from a cell in a matrix, to its column, then to its >> > stateset definition to infer that something is a dna sequence >> > alignment (say) >> > >> > * matrices are far, far too bulky in the granular serialization that >> > is used. Each cell in a matrix is described by 4 triples. Assuming >> > even a small-ish alignment (20 taxa, 2000 characters), we'd have to >> > use 160,000 triples to describe that. TreeBASE would need triple >> > stores of the scale that UniProt is using. >> > >> > If we want to use CDAO as part of a semantic web stack (for HIP, >> > TreeBASE3, etc.) it needs to be developed further. >> > >> > Rutger >> > >> > -- >> > Dr. Rutger A. Vos >> > School of Biological Sciences >> > Philip Lyle Building, Level 4 >> > University of Reading >> > Reading, RG6 6BX, United Kingdom >> > Tel: +44 (0) 118 378 7535 >> > http://rutgervos.blogspot.com >> > > > > > -- > Dr. Rutger A. Vos > School of Biological Sciences > Philip Lyle Building, Level 4 > University of Reading > Reading, RG6 6BX, United Kingdom > Tel: +44 (0) 118 378 7535 > http://rutgervos.blogspot.com > > ------------------------------------------------------------------------------ > BlackBerry® DevCon Americas, Oct. 18-20, San Francisco, CA > Learn about the latest advances in developing for the > BlackBerry® mobile platform with sessions, labs & more. > See new tools and technologies. Register for BlackBerry® DevCon today! > http://p.sf.net/sfu/rim-devcon-copy1 > _______________________________________________ > Nexml-discuss mailing list > Nex...@li... > https://lists.sourceforge.net/lists/listinfo/nexml-discuss > > ---------------------------------------------------------------------------- -- Doing More with Less: The Next Generation Virtual Desktop What are the key obstacles that have prevented many mid-market businesses from deploying virtual desktops? How do next-generation virtual desktops provide companies an easier-to-deploy, easier-to-manage and more affordable virtual desktop model.http://www.accelacomm.com/jaw/sfnl/114/51426474/______________________ _________________________ CDAO-discuss mailing list CDA...@li... https://lists.sourceforge.net/lists/listinfo/cdao-discuss |
|
From: Hilmar L. <hl...@ne...> - 2011-09-16 15:19:54
|
I think this is a good opportunity for all of us to realize that CDAO at this point is like most other open-source, open development projects. It will thrive, evolve, and improve on people volunteering their time and contribution, i.e., being developers, rather than from complaining or pleading to some assumed developers out there. I realize that this is different from the gatekeeper model frequently employed among the OBO ontologies, but CDAO isn't an OBO ontology, nor is there a funded project right now that has some degree of dedicated support for it. And without that, the gatekeeper model simply doesn't work. So while I agree with the importance of CDAO's role, its going to take contributions from everyone who has a need for its use, and securing funding to sustain CDAO's continued development. Jamie - it'd indeed be great if you can help out. Let us know if there are some ways we can help with your grant proposal ideas. -hilmar On Sep 14, 2011, at 6:06 AM, Rutger Vos wrote: > Hi all, > > it seems to me that it is critically important for a number of > projects we are all involved in (HIP, MIAPA, TreeBASE, etc.) that > there is an ontology for evolutionary informatics whose developers > respond to user needs. I dearly hope someone will step forward to push > further CDAO development. > > If not, I guess it might mean we will have to resort to a combination > of EDAM, MAO, and other ontologies - although those might not respond > as much to user requests as we need at this point in the game. > > Rutger > > On Tue, Sep 6, 2011 at 11:41 AM, Rutger Vos <R....@re...> > wrote: >> Hi all, >> >> I hate to be blunt, but I'm worried about CDAO: it's broken beyond >> the >> point of usability and it doesn't seem to be maintained. I've now >> tried to do some serious work with it, querying RDF graphs to fish >> out >> the taxa, character state matrices and trees and it is simply not >> working correctly. Some examples: >> >> * characters (i.e. columns) do have an ordering that needs to be >> maintained, especially for molecular data, this should either be done >> using column numbers or using an rdf list in the character >> specification. (Arguably, orderings in other things - lists of taxa, >> trees - would be useful too.) >> >> * it needs to be discoverable what data type a matrix is. It is no >> good having to go from a cell in a matrix, to its column, then to its >> stateset definition to infer that something is a dna sequence >> alignment (say) >> >> * matrices are far, far too bulky in the granular serialization that >> is used. Each cell in a matrix is described by 4 triples. Assuming >> even a small-ish alignment (20 taxa, 2000 characters), we'd have to >> use 160,000 triples to describe that. TreeBASE would need triple >> stores of the scale that UniProt is using. >> >> If we want to use CDAO as part of a semantic web stack (for HIP, >> TreeBASE3, etc.) it needs to be developed further. >> >> Rutger >> >> -- >> Dr. Rutger A. Vos >> School of Biological Sciences >> Philip Lyle Building, Level 4 >> University of Reading >> Reading, RG6 6BX, United Kingdom >> Tel: +44 (0) 118 378 7535 >> http://rutgervos.blogspot.com >> > > > > -- > Dr. Rutger A. Vos > School of Biological Sciences > Philip Lyle Building, Level 4 > University of Reading > Reading, RG6 6BX, United Kingdom > Tel: +44 (0) 118 378 7535 > http://rutgervos.blogspot.com -- =========================================================== : Hilmar Lapp -:- Durham, NC -:- informatics.nescent.org : =========================================================== |
|
From: Christian Z. <cmz...@ya...> - 2011-09-15 17:55:33
|
Hi, Rutger: To be honest, I have not been following CDAO, so I am not entirely clear what it encompasses. In any case, if you think my expertise on molecular evolution and comparative genomics could be useful in revitalizing CDAO, I'd be happy to contribute. Christian >________________________________ >From: Rutger Vos <R....@re...> >To: CDAO list <cda...@li...>; NeXML-discuss (list) <nex...@li...>; Hilmar Lapp <hl...@ne...>; MIAPA <mia...@go...> >Sent: Wednesday, September 14, 2011 3:06 AM >Subject: Re: [Nexml-discuss] CDAO status and future > >Hi all, > >it seems to me that it is critically important for a number of >projects we are all involved in (HIP, MIAPA, TreeBASE, etc.) that >there is an ontology for evolutionary informatics whose developers >respond to user needs. I dearly hope someone will step forward to push >further CDAO development. > >If not, I guess it might mean we will have to resort to a combination >of EDAM, MAO, and other ontologies - although those might not respond >as much to user requests as we need at this point in the game. > >Rutger > >On Tue, Sep 6, 2011 at 11:41 AM, Rutger Vos <R....@re...> wrote: >> Hi all, >> >> I hate to be blunt, but I'm worried about CDAO: it's broken beyond the >> point of usability and it doesn't seem to be maintained. I've now >> tried to do some serious work with it, querying RDF graphs to fish out >> the taxa, character state matrices and trees and it is simply not >> working correctly. Some examples: >> >> * characters (i.e. columns) do have an ordering that needs to be >> maintained, especially for molecular data, this should either be done >> using column numbers or using an rdf list in the character >> specification. (Arguably, orderings in other things - lists of taxa, >> trees - would be useful too.) >> >> * it needs to be discoverable what data type a matrix is. It is no >> good having to go from a cell in a matrix, to its column, then to its >> stateset definition to infer that something is a dna sequence >> alignment (say) >> >> * matrices are far, far too bulky in the granular serialization that >> is used. Each cell in a matrix is described by 4 triples. Assuming >> even a small-ish alignment (20 taxa, 2000 characters), we'd have to >> use 160,000 triples to describe that. TreeBASE would need triple >> stores of the scale that UniProt is using. >> >> If we want to use CDAO as part of a semantic web stack (for HIP, >> TreeBASE3, etc.) it needs to be developed further. >> >> Rutger >> >> -- >> Dr. Rutger A. Vos >> School of Biological Sciences >> Philip Lyle Building, Level 4 >> University of Reading >> Reading, RG6 6BX, United Kingdom >> Tel: +44 (0) 118 378 7535 >> http://rutgervos.blogspot.com >> > > > >-- >Dr. Rutger A. Vos >School of Biological Sciences >Philip Lyle Building, Level 4 >University of Reading >Reading, RG6 6BX, United Kingdom >Tel: +44 (0) 118 378 7535 >http://rutgervos.blogspot.com > >------------------------------------------------------------------------------ >BlackBerry® DevCon Americas, Oct. 18-20, San Francisco, CA >Learn about the latest advances in developing for the >BlackBerry® mobile platform with sessions, labs & more. >See new tools and technologies. Register for BlackBerry® DevCon today! >http://p.sf.net/sfu/rim-devcon-copy1 >_______________________________________________ >Nexml-discuss mailing list >Nex...@li... >https://lists.sourceforge.net/lists/listinfo/nexml-discuss > > > |
|
From: Rutger V. <R....@re...> - 2011-09-14 15:19:38
|
Hi Jamie, I think it would be terrific if you were to get involved with CDAO! Rutger On Wed, Sep 14, 2011 at 5:06 PM, Jamie Estill <jam...@gm...> wrote: > I would be able to help with CDAO development and would be willing to help > apply for funding for future development of CDAO. I'm a postdoc with Jim > Leebens-Mack at UGA and have an adjunct position at the University of > Cincinnati. I am finishing up an OWL ontology for gene tree to species tree > reconciliation but I'm fairly new to this compared to some others. > -- Jamie > On Wed, Sep 14, 2011 at 6:06 AM, Rutger Vos <R....@re...> wrote: >> >> Hi all, >> >> it seems to me that it is critically important for a number of >> projects we are all involved in (HIP, MIAPA, TreeBASE, etc.) that >> there is an ontology for evolutionary informatics whose developers >> respond to user needs. I dearly hope someone will step forward to push >> further CDAO development. >> >> If not, I guess it might mean we will have to resort to a combination >> of EDAM, MAO, and other ontologies - although those might not respond >> as much to user requests as we need at this point in the game. >> >> Rutger >> >> On Tue, Sep 6, 2011 at 11:41 AM, Rutger Vos <R....@re...> wrote: >> > Hi all, >> > >> > I hate to be blunt, but I'm worried about CDAO: it's broken beyond the >> > point of usability and it doesn't seem to be maintained. I've now >> > tried to do some serious work with it, querying RDF graphs to fish out >> > the taxa, character state matrices and trees and it is simply not >> > working correctly. Some examples: >> > >> > * characters (i.e. columns) do have an ordering that needs to be >> > maintained, especially for molecular data, this should either be done >> > using column numbers or using an rdf list in the character >> > specification. (Arguably, orderings in other things - lists of taxa, >> > trees - would be useful too.) >> > >> > * it needs to be discoverable what data type a matrix is. It is no >> > good having to go from a cell in a matrix, to its column, then to its >> > stateset definition to infer that something is a dna sequence >> > alignment (say) >> > >> > * matrices are far, far too bulky in the granular serialization that >> > is used. Each cell in a matrix is described by 4 triples. Assuming >> > even a small-ish alignment (20 taxa, 2000 characters), we'd have to >> > use 160,000 triples to describe that. TreeBASE would need triple >> > stores of the scale that UniProt is using. >> > >> > If we want to use CDAO as part of a semantic web stack (for HIP, >> > TreeBASE3, etc.) it needs to be developed further. >> > >> > Rutger >> > >> > -- >> > Dr. Rutger A. Vos >> > School of Biological Sciences >> > Philip Lyle Building, Level 4 >> > University of Reading >> > Reading, RG6 6BX, United Kingdom >> > Tel: +44 (0) 118 378 7535 >> > http://rutgervos.blogspot.com >> > >> >> >> >> -- >> Dr. Rutger A. Vos >> School of Biological Sciences >> Philip Lyle Building, Level 4 >> University of Reading >> Reading, RG6 6BX, United Kingdom >> Tel: +44 (0) 118 378 7535 >> http://rutgervos.blogspot.com >> >> -- >> You received this message because you are subscribed to the Google >> Groups "MIAPA" group. >> For more options, visit this group at >> http://groups.google.com/group/miapa-discuss?hl=en > > > > -- > ----------------------------------------- > James C. Estill > Jam...@gm... > http://jestill.myweb.uga.edu > ----------------------------------------- > > ------------------------------------------------------------------------------ > BlackBerry® DevCon Americas, Oct. 18-20, San Francisco, CA > Learn about the latest advances in developing for the > BlackBerry® mobile platform with sessions, labs & more. > See new tools and technologies. Register for BlackBerry® DevCon today! > http://p.sf.net/sfu/rim-devcon-copy1 > _______________________________________________ > Nexml-discuss mailing list > Nex...@li... > https://lists.sourceforge.net/lists/listinfo/nexml-discuss > > -- Dr. Rutger A. Vos School of Biological Sciences Philip Lyle Building, Level 4 University of Reading Reading, RG6 6BX, United Kingdom Tel: +44 (0) 118 378 7535 http://rutgervos.blogspot.com |
|
From: Jamie E. <jam...@gm...> - 2011-09-14 15:06:48
|
I would be able to help with CDAO development and would be willing to help apply for funding for future development of CDAO. I'm a postdoc with Jim Leebens-Mack at UGA and have an adjunct position at the University of Cincinnati. I am finishing up an OWL ontology for gene tree to species tree reconciliation but I'm fairly new to this compared to some others. -- Jamie On Wed, Sep 14, 2011 at 6:06 AM, Rutger Vos <R....@re...> wrote: > Hi all, > > it seems to me that it is critically important for a number of > projects we are all involved in (HIP, MIAPA, TreeBASE, etc.) that > there is an ontology for evolutionary informatics whose developers > respond to user needs. I dearly hope someone will step forward to push > further CDAO development. > > If not, I guess it might mean we will have to resort to a combination > of EDAM, MAO, and other ontologies - although those might not respond > as much to user requests as we need at this point in the game. > > Rutger > > On Tue, Sep 6, 2011 at 11:41 AM, Rutger Vos <R....@re...> wrote: > > Hi all, > > > > I hate to be blunt, but I'm worried about CDAO: it's broken beyond the > > point of usability and it doesn't seem to be maintained. I've now > > tried to do some serious work with it, querying RDF graphs to fish out > > the taxa, character state matrices and trees and it is simply not > > working correctly. Some examples: > > > > * characters (i.e. columns) do have an ordering that needs to be > > maintained, especially for molecular data, this should either be done > > using column numbers or using an rdf list in the character > > specification. (Arguably, orderings in other things - lists of taxa, > > trees - would be useful too.) > > > > * it needs to be discoverable what data type a matrix is. It is no > > good having to go from a cell in a matrix, to its column, then to its > > stateset definition to infer that something is a dna sequence > > alignment (say) > > > > * matrices are far, far too bulky in the granular serialization that > > is used. Each cell in a matrix is described by 4 triples. Assuming > > even a small-ish alignment (20 taxa, 2000 characters), we'd have to > > use 160,000 triples to describe that. TreeBASE would need triple > > stores of the scale that UniProt is using. > > > > If we want to use CDAO as part of a semantic web stack (for HIP, > > TreeBASE3, etc.) it needs to be developed further. > > > > Rutger > > > > -- > > Dr. Rutger A. Vos > > School of Biological Sciences > > Philip Lyle Building, Level 4 > > University of Reading > > Reading, RG6 6BX, United Kingdom > > Tel: +44 (0) 118 378 7535 > > http://rutgervos.blogspot.com > > > > > > -- > Dr. Rutger A. Vos > School of Biological Sciences > Philip Lyle Building, Level 4 > University of Reading > Reading, RG6 6BX, United Kingdom > Tel: +44 (0) 118 378 7535 > http://rutgervos.blogspot.com > > -- > You received this message because you are subscribed to the Google > Groups "MIAPA" group. > For more options, visit this group at > http://groups.google.com/group/miapa-discuss?hl=en > -- ----------------------------------------- James C. Estill Jam...@gm... http://jestill.myweb.uga.edu ----------------------------------------- |
|
From: Rutger V. <R....@re...> - 2011-09-14 10:06:10
|
Hi all, it seems to me that it is critically important for a number of projects we are all involved in (HIP, MIAPA, TreeBASE, etc.) that there is an ontology for evolutionary informatics whose developers respond to user needs. I dearly hope someone will step forward to push further CDAO development. If not, I guess it might mean we will have to resort to a combination of EDAM, MAO, and other ontologies - although those might not respond as much to user requests as we need at this point in the game. Rutger On Tue, Sep 6, 2011 at 11:41 AM, Rutger Vos <R....@re...> wrote: > Hi all, > > I hate to be blunt, but I'm worried about CDAO: it's broken beyond the > point of usability and it doesn't seem to be maintained. I've now > tried to do some serious work with it, querying RDF graphs to fish out > the taxa, character state matrices and trees and it is simply not > working correctly. Some examples: > > * characters (i.e. columns) do have an ordering that needs to be > maintained, especially for molecular data, this should either be done > using column numbers or using an rdf list in the character > specification. (Arguably, orderings in other things - lists of taxa, > trees - would be useful too.) > > * it needs to be discoverable what data type a matrix is. It is no > good having to go from a cell in a matrix, to its column, then to its > stateset definition to infer that something is a dna sequence > alignment (say) > > * matrices are far, far too bulky in the granular serialization that > is used. Each cell in a matrix is described by 4 triples. Assuming > even a small-ish alignment (20 taxa, 2000 characters), we'd have to > use 160,000 triples to describe that. TreeBASE would need triple > stores of the scale that UniProt is using. > > If we want to use CDAO as part of a semantic web stack (for HIP, > TreeBASE3, etc.) it needs to be developed further. > > Rutger > > -- > Dr. Rutger A. Vos > School of Biological Sciences > Philip Lyle Building, Level 4 > University of Reading > Reading, RG6 6BX, United Kingdom > Tel: +44 (0) 118 378 7535 > http://rutgervos.blogspot.com > -- Dr. Rutger A. Vos School of Biological Sciences Philip Lyle Building, Level 4 University of Reading Reading, RG6 6BX, United Kingdom Tel: +44 (0) 118 378 7535 http://rutgervos.blogspot.com |
|
From: Arlin S. <ar...@um...> - 2011-09-09 14:09:46
|
I could respond to some of those points, by explaining why CDAO does things a particular way, but that would not address the bigger issue of how the CDAO project (which-- given that we do not have funding-- is nothing more or less than the people on this list) is going to respond to this opportunity, or any other opportunity to make CDAO actually useful. Arlin On Sep 6, 2011, at 5:41 AM, Rutger Vos wrote: > Hi all, > > I hate to be blunt, but I'm worried about CDAO: it's broken beyond the > point of usability and it doesn't seem to be maintained. I've now > tried to do some serious work with it, querying RDF graphs to fish out > the taxa, character state matrices and trees and it is simply not > working correctly. Some examples: > > * characters (i.e. columns) do have an ordering that needs to be > maintained, especially for molecular data, this should either be done > using column numbers or using an rdf list in the character > specification. (Arguably, orderings in other things - lists of taxa, > trees - would be useful too.) > > * it needs to be discoverable what data type a matrix is. It is no > good having to go from a cell in a matrix, to its column, then to its > stateset definition to infer that something is a dna sequence > alignment (say) > > * matrices are far, far too bulky in the granular serialization that > is used. Each cell in a matrix is described by 4 triples. Assuming > even a small-ish alignment (20 taxa, 2000 characters), we'd have to > use 160,000 triples to describe that. TreeBASE would need triple > stores of the scale that UniProt is using. > > If we want to use CDAO as part of a semantic web stack (for HIP, > TreeBASE3, etc.) it needs to be developed further. > > Rutger > > -- > Dr. Rutger A. Vos > School of Biological Sciences > Philip Lyle Building, Level 4 > University of Reading > Reading, RG6 6BX, United Kingdom > Tel: +44 (0) 118 378 7535 > http://rutgervos.blogspot.com > > ------------------------------------------------------------------------------ > Special Offer -- Download ArcSight Logger for FREE! > Finally, a world-class log management solution at an even better > price-free! And you'll get a free "Love Thy Logs" t-shirt when you > download Logger. Secure your free ArcSight Logger TODAY! > http://p.sf.net/sfu/arcsisghtdev2dev > _______________________________________________ > CDAO-discuss mailing list > CDA...@li... > https://lists.sourceforge.net/lists/listinfo/cdao-discuss ------- Arlin Stoltzfus (ar...@um...) Fellow, IBBR; Adj. Assoc. Prof., UMCP; Research Biologist, NIST IBBR, 9600 Gudelsky Drive, Rockville, MD tel: 240 314 6208; web: www.molevol.org |
|
From: Rutger V. <R....@re...> - 2011-09-06 09:41:18
|
Hi all, I hate to be blunt, but I'm worried about CDAO: it's broken beyond the point of usability and it doesn't seem to be maintained. I've now tried to do some serious work with it, querying RDF graphs to fish out the taxa, character state matrices and trees and it is simply not working correctly. Some examples: * characters (i.e. columns) do have an ordering that needs to be maintained, especially for molecular data, this should either be done using column numbers or using an rdf list in the character specification. (Arguably, orderings in other things - lists of taxa, trees - would be useful too.) * it needs to be discoverable what data type a matrix is. It is no good having to go from a cell in a matrix, to its column, then to its stateset definition to infer that something is a dna sequence alignment (say) * matrices are far, far too bulky in the granular serialization that is used. Each cell in a matrix is described by 4 triples. Assuming even a small-ish alignment (20 taxa, 2000 characters), we'd have to use 160,000 triples to describe that. TreeBASE would need triple stores of the scale that UniProt is using. If we want to use CDAO as part of a semantic web stack (for HIP, TreeBASE3, etc.) it needs to be developed further. Rutger -- Dr. Rutger A. Vos School of Biological Sciences Philip Lyle Building, Level 4 University of Reading Reading, RG6 6BX, United Kingdom Tel: +44 (0) 118 378 7535 http://rutgervos.blogspot.com |
|
From: Hilmar L. <hl...@ne...> - 2011-08-23 21:31:09
|
Doesn't SF allow for serving files off of a directory? We could then drop a .htaccess in that folder that maps .owl to mime-type application/rdf+xml. Rutger - I don't think the problem is that reasoners can't figure out the serialization (the extension could be used for that as well), but that they think it's not an OWL file to start with. Same result, anyway. -hilmar On Aug 24, 2011, at 4:51 AM, Arlin Stoltzfus wrote: > On Aug 23, 2011, at 3:53 AM, Rutger Vos wrote: > >> Hi all, >> >> the link that is resolved when some client (reasoner, whatever) >> follows http://www.evolutionaryontology.org/cdao/1.0/cdao.owl leads >> to >> the cvs version of the ontology. This file is served up with the >> mime-type "Content-Type: text/plain; charset=UTF-8". >> >> This makes it essentially useless because any client looking to >> consume the ontology and use it to reason over now can't determine >> what RDF serialization (xml/turtle/n3) it is. This needs to be >> changed >> so that the mime-type becomes application/rdf+xml; charset=UTF-8. >> >> The way to do this is to serve a stable version on the web site >> itself >> (not on sourceforge) so that you can control these things (cvs >> doesn't >> allow you to specify mime types, svn does though). Then you create an >> .htaccess file in the folder where the ontology resides and set the >> mime type for .owl files to application/rdf+xml > > Ugh. This is the kind of thing that it takes me hours to figure out. > Previously I was getting help from NESCent IT on this. > > Right now, there is no web site other than sourceforge, so we can't do > it this way. We used to have a separate web site, but we decided that > we were relying on sourceforge anyway, and it would be better to move > everything to one site where we controlled permissions. > >> Alternatively, you can redirect to a little perl cgi script that >> pulls >> in the cvs version and spits it out with the correct CGI header. > > By "alternatively", you mean that we can do this on sourceforge? If > so, then go ahead and do it. I can make sure that you have the right > permissions. You will need to have shell access to do this. > >> This has come up before (in October 2009), back when the mime-type >> was >> application/xml. Not sure why it was since changed to text/plain. > > Probably happened when we migrated to sourceforge-only, in 2010. > > Arlin > ------- > Arlin Stoltzfus (ar...@um...) > Fellow, IBBR; Adj. Assoc. Prof., UMCP; Research Biologist, NIST > IBBR, 9600 Gudelsky Drive, Rockville, MD > tel: 240 314 6208; web: www.molevol.org > > > ------------------------------------------------------------------------------ > Get a FREE DOWNLOAD! and learn more about uberSVN rich system, > user administration capabilities and model configuration. Take > the hassle out of deploying and managing Subversion and the > tools developers use with it. http://p.sf.net/sfu/wandisco-d2d-2 > _______________________________________________ > CDAO-discuss mailing list > CDA...@li... > https://lists.sourceforge.net/lists/listinfo/cdao-discuss -- =========================================================== : Hilmar Lapp -:- Durham, NC -:- informatics.nescent.org : =========================================================== |
|
From: Arlin S. <ar...@um...> - 2011-08-23 19:52:13
|
On Aug 23, 2011, at 3:53 AM, Rutger Vos wrote: > Hi all, > > the link that is resolved when some client (reasoner, whatever) > follows http://www.evolutionaryontology.org/cdao/1.0/cdao.owl leads to > the cvs version of the ontology. This file is served up with the > mime-type "Content-Type: text/plain; charset=UTF-8". > > This makes it essentially useless because any client looking to > consume the ontology and use it to reason over now can't determine > what RDF serialization (xml/turtle/n3) it is. This needs to be changed > so that the mime-type becomes application/rdf+xml; charset=UTF-8. > > The way to do this is to serve a stable version on the web site itself > (not on sourceforge) so that you can control these things (cvs doesn't > allow you to specify mime types, svn does though). Then you create an > .htaccess file in the folder where the ontology resides and set the > mime type for .owl files to application/rdf+xml Ugh. This is the kind of thing that it takes me hours to figure out. Previously I was getting help from NESCent IT on this. Right now, there is no web site other than sourceforge, so we can't do it this way. We used to have a separate web site, but we decided that we were relying on sourceforge anyway, and it would be better to move everything to one site where we controlled permissions. > Alternatively, you can redirect to a little perl cgi script that pulls > in the cvs version and spits it out with the correct CGI header. By "alternatively", you mean that we can do this on sourceforge? If so, then go ahead and do it. I can make sure that you have the right permissions. You will need to have shell access to do this. > This has come up before (in October 2009), back when the mime-type was > application/xml. Not sure why it was since changed to text/plain. Probably happened when we migrated to sourceforge-only, in 2010. Arlin ------- Arlin Stoltzfus (ar...@um...) Fellow, IBBR; Adj. Assoc. Prof., UMCP; Research Biologist, NIST IBBR, 9600 Gudelsky Drive, Rockville, MD tel: 240 314 6208; web: www.molevol.org |
|
From: Rutger V. <R....@re...> - 2011-08-23 07:53:41
|
Hi all, the link that is resolved when some client (reasoner, whatever) follows http://www.evolutionaryontology.org/cdao/1.0/cdao.owl leads to the cvs version of the ontology. This file is served up with the mime-type "Content-Type: text/plain; charset=UTF-8". This makes it essentially useless because any client looking to consume the ontology and use it to reason over now can't determine what RDF serialization (xml/turtle/n3) it is. This needs to be changed so that the mime-type becomes application/rdf+xml; charset=UTF-8. The way to do this is to serve a stable version on the web site itself (not on sourceforge) so that you can control these things (cvs doesn't allow you to specify mime types, svn does though). Then you create an .htaccess file in the folder where the ontology resides and set the mime type for .owl files to application/rdf+xml Alternatively, you can redirect to a little perl cgi script that pulls in the cvs version and spits it out with the correct CGI header. This has come up before (in October 2009), back when the mime-type was application/xml. Not sure why it was since changed to text/plain. Thanks, Rutger -- Dr. Rutger A. Vos School of Biological Sciences Philip Lyle Building, Level 4 University of Reading Reading, RG6 6BX, United Kingdom Tel: +44 (0) 118 378 7535 http://rutgervos.blogspot.com |
|
From: Jim B. <ba...@ne...> - 2011-08-19 17:44:30
|
Hi Arlin, On Aug 19, 2011, at 10:42 AM, Arlin Stoltzfus wrote: > I am all in favor of "saving some trouble making decisions on > mechanical details"! But I'm not exactly sure what you mean. We > developed CDAO using most of the same principles that OBO used to list > as "principles" on its web site. But these "principles" are high- > level, e.g., the principle of giving everything a URIs does not say > anything other than to give every class and relation a URI. Recently > OBO has listed some naming conventions as a "principle" (the kind of > category error that one would have expected ontologists to avoid). > What conventions are you talking about, exactly? I am not really talking about the design principles, just more mundane things. One example is URI conventions like these: https://docs.google.com/Doc?docid=0Acx6Blq96uycZHpwcm5td18wZGtkMmdiZ3Y&hl=en_US Standard ways to formulate PURLs for ontology terms, file releases, homepages, etc. Thanks, Jim ____________________________________________ James P. Balhoff, Ph.D. National Evolutionary Synthesis Center 2024 West Main St., Suite A200 Durham, NC 27705 USA |
|
From: Hilmar L. <hl...@ne...> - 2011-08-19 14:54:03
|
As inconvenient as it can sometimes be when staring at OBO identifiers in OWL files, the troubles that come with wanting to change the label when the label is also the identifier is easily worse, so I'd support a move towards the OBO standard URI scheme. Having said that, this issue has just recently been the subject of a long and contested debate on the HCLS or Linked Data W3C list (I forget which), so the issue is controversial among different communities. -hilmar Sent with a tap. On Aug 18, 2011, at 3:15 PM, Jim Balhoff <ba...@ne...> wrote: > A more administrative question is what folks' thoughts are on becoming more compliant with OBO foundry conventions (opaque identifiers, standard URI scheme for releases). I think there are some benefits to this as far as more interaction with other ontology communities and also adoption of conventions which can save some trouble making decisions on mechanical details. |
|
From: Arlin S. <ar...@um...> - 2011-08-19 14:42:34
|
On Aug 18, 2011, at 3:15 PM, Jim Balhoff wrote:
> Since I have been working with CDAO quite a bit lately, I am
> interested in taking on the project management. Part of your message
> describes pretty well what I've been doing:
Thanks for responding with your interest!
> As I use CDAO to represent semantically annotated comparative data,
> I have been trying to work out how a data matrix is related to "real-
> world" organisms and taxa. I view instances of CDAO matrix-related
> classes as information content entities which can denote organisms
> and their features. So now I have some accessory properties and
> property chains which can propagate phenotype annotations from
> states onto the things that are denoted by OTUs. CDAO would be a
> good place to add these properties. I wonder if there is any
> controversy about whether the instances of CDAO classes are
> information artifacts (my current view) or actually organisms and
> their characteristics. Has CDAO taken a stance on this before?
CDAO does not explicitly take a stance, and the issue has not been
discussed much. My naive view is that instances of CDAO are
information artefacts defined by their instrumental role in
comparative analysis. This is obviously true for a "character matrix"
and is arguably true for phylogenetic trees. To the extent that a
tree is a kind of graph, it is an information artefact.
Possibly there is a logical problem when we include "data" (e.g.,
"character data"). I think some OBO people pointed this out when
Hilmar presented CDAO at an OBO meeting 2 years ago. The resolution
to this might be to note that the character "state", e.g., "3" or
"present" or "T" does not directly represent an observational fact,
but represents an encoding of an interpretation of biological
observations relative to other observations. An indication of this is
that we accept that two different things can have the "same" state,
e.g., in a NEXUS file we might have a token that represents a given
state of a character ("reduced"), and that token could be assigned to
OTUs 2, 3, 5 and 18, even though OTUs 2, 3, 5 and 18 are not the same
things, and even though the set of observations (facts) that underly
the "reduced" designation are not the same things. And this also
would be clear in the ways that computer programs analyze the "states".
Apparently, "is_about" is the magic predicate that bridges the divide
between information artefacts and the real world, i.e., the character
states are "about" some biological entity. However, I don't claim to
fully understand the issues.
> CDAO has been very useful as a framework for recording phenotype
> annotations in a logically consistent way across both Phenoscape and
> the Hymenoptera anatomy ontology project (the other group I'm
> working with).
I'm excited about the prospect of working out problems in CDAO in the
context of actual uses.
> A more administrative question is what folks' thoughts are on
> becoming more compliant with OBO foundry conventions (opaque
> identifiers, standard URI scheme for releases). I think there are
> some benefits to this as far as more interaction with other ontology
> communities and also adoption of conventions which can save some
> trouble making decisions on mechanical details.
I am all in favor of "saving some trouble making decisions on
mechanical details"! But I'm not exactly sure what you mean. We
developed CDAO using most of the same principles that OBO used to list
as "principles" on its web site. But these "principles" are high-
level, e.g., the principle of giving everything a URIs does not say
anything other than to give every class and relation a URI. Recently
OBO has listed some naming conventions as a "principle" (the kind of
category error that one would have expected ontologists to avoid).
What conventions are you talking about, exactly?
Arlin
-------
Arlin Stoltzfus (ar...@um...)
Fellow, IBBR; Adj. Assoc. Prof., UMCP; Research Biologist, NIST
IBBR, 9600 Gudelsky Drive, Rockville, MD
tel: 240 314 6208; web: www.molevol.org
|
|
From: Jim B. <ba...@ne...> - 2011-08-18 19:15:43
|
Hi Arlin, Since I have been working with CDAO quite a bit lately, I am interested in taking on the project management. Part of your message describes pretty well what I've been doing: On Jul 25, 2011, at 1:55 PM, Arlin Stoltzfus wrote: > The current version of CDAO is a bit bloated and messy, in my > opinion. Enrico and others have suggested splitting it into an > ontology for trees, and an ontology for comparative data, knitted > together in a larger ontology. CDAO does not take a clear or > meaningful position on some upper-level ontology issues, which are > very complex for comparative data. Once upon a time, I imagined that > such representation issues would be worked out by developing critical > tests based on use-cases. This hasn't happened for us-- we have > benefited from some people trying to use CDAO, but mainly in terms of > very low-level technical details rather than high-level concepts. As I use CDAO to represent semantically annotated comparative data, I have been trying to work out how a data matrix is related to "real-world" organisms and taxa. I view instances of CDAO matrix-related classes as information content entities which can denote organisms and their features. So now I have some accessory properties and property chains which can propagate phenotype annotations from states onto the things that are denoted by OTUs. CDAO would be a good place to add these properties. I wonder if there is any controversy about whether the instances of CDAO classes are information artifacts (my current view) or actually organisms and their characteristics. Has CDAO taken a stance on this before? CDAO has been very useful as a framework for recording phenotype annotations in a logically consistent way across both Phenoscape and the Hymenoptera anatomy ontology project (the other group I'm working with). A more administrative question is what folks' thoughts are on becoming more compliant with OBO foundry conventions (opaque identifiers, standard URI scheme for releases). I think there are some benefits to this as far as more interaction with other ontology communities and also adoption of conventions which can save some trouble making decisions on mechanical details. Thanks, Jim ____________________________________________ James P. Balhoff, Ph.D. National Evolutionary Synthesis Center 2024 West Main St., Suite A200 Durham, NC 27705 USA |
|
From: Nico C. <nce...@fl...> - 2011-08-15 15:54:13
|
TDWG MIAPA Workshop Call For Participation: Steps towards a Minimum Information About a Phylogenetic Analysis (MIAPA) Standard Synopsis Many phylogenetic analysis results are published in ways that present serious barriers to their reuse in numerous research applications that would stand to benefit from them. While some of these barriers are well understood, such as issues with adherence to standard exchange formats, those centering on the associated metadata necessary for researchers to evaluate or reuse a published phylogeny have only recently begun to be articulated. One of the critical next steps towards formalizing these metadata requirements as a minimum reporting standard is to convene meetings of key stakeholder communities with the goal to identify information attributes necessary and desirable for facilitating reuse, and to build consensus on their priority. To this end, we are holding a workshop at the 2011 Biodiversity Information Standards (TDWG) Conference to determine how a future reporting standard for phylogenetic analyses can best serve biodiversity science and related research applications. We invite all interested colleagues to participate. Background The workshop of the Biodiversity Information Standards (TDWG) Phylogenetics Standards Interest Group held at the 2010 TDWG conference included a project focused on how to publish re-usable trees that can be linked into an emerging global web of data. Through follow-up work, this led to the following tangible results: An online draft report of the 2010 TDWG workshop [1], and a corresponding manuscript on best practices for publishing phylogenetic trees (Stoltzfus et al. in preparation); An 2011 iEvoBio presentation on “Publishing re-usable phylogenetic trees, in theory and in practice” [2]; A lighting talk presentation and Birds-of-a-Feather gathering at 2011 iEvoBio, and A survey group that explored barriers to re-use and developed plans for a survey These activities have considerably clarified our understanding of the theory and practice of publishing re-usable phylogenetic trees: how many phylogenies are published each year, the (low) frequency of archiving, what archives and tools are available, what policies are in force, etc. We have identified a number of barriers to re-use involving such aspects as technology, standards, culture, and access. Many of these barriers can be interpreted as a consequence of the lack of a community-agreed standard for what constitutes a well documented phylogenetic record. In the absence of such a standard, trees are often archived as image files rather than in appropriate data exchange formats, and lack important accompanying information (metadata), such as externally meaningful identifiers, that would be needed to make them useful to others. The idea of a Minimum Information About a Phylogenetic Analysis (MIAPA) standard has been suggested [3], but so far there has not been a deliberate process to develop and disseminate a community standard. Meanwhile, a number of systematics and evolution journals have begun to require archiving of the data underlying published research findings [4]. The emerging cultural shift in data archiving and sharing promoted by this policy change offers a unique window of opportunity to move ahead with the development and actual specification of a MIAPA standard. Similar to other minimum reporting standards [5], the primary focus of a future MIAPA standard would be on defining a “checklist” of metadata information attributes that, at a minimum, needs to accompany an archived phylogenetic analysis, and to which standards values for these attributes would need to adhere. The key step in developing community consensus on these elements of the standard is to convene a series of meetings that collectively involve participants from all major groups of stakeholders who would be affected by such a standard, such as users, producers, publishers, or archivists of phylogenetic analyses. To aid this process, the Phylogenetics Standards Interest Group is holding a workshop at the 2011 TDWG conference, with the goal to obtain consensus requirements and priorities for a MIAPA checklist for the purposes of biodiversity science, taxonomy, museum collections, and related research applications. Goals and deliverables The main goal of the workshop is to develop a shared understanding of the role that a MIAPA standard could play in facilitating re-use of phylogenetic analyses for the biodiversity science and related communities, and what the standard would need to specify in order to best fill that role. Possible deliverables include A draft set of information attributes that should or could be included in a provisional MIAPA checklist, with a level of consensus for each of them. A database with use-cases based on exemplifying publications, that report phylogenies to elucidate a broad spectrum of questions relating to biodiversity science. A refined MIAPA survey to be informed by biodiversity science cases for reuse. A plan for further community engagement and consensus-building among biodiversity science stakeholders. Workshop format The workshop will start with a few presentations focused on (i) introducing MIAPA and its potential in facilitating reuse (J. Leebens-Mack); (ii) summarizing recent developments and current status of MIAPA-related efforts (A. Stoltzfus); and (iii) past experiences and resulting best practice recommendations on developing a minimum reporting checklist standard (D. Field). The rest of the workshop will be hands-on. Participants in the workshop will break out into groups to address separate issues according to the anticipated deliverables and best practice recommendations. The workshop will be 1.5 days in duration, and be held during the 2011 Biodiversity Information Standards (TDWG) conference, to take place Oct 17 to 21, 2011 in New Orleans, USA. (http://www.tdwg.org/conference2011/). The workshop will start in the afternoon of Monday, Oct 17, and end on Tuesday. Oct 18. How to participate Participation in the workshop is open to everyone interested. However, space is limited, and we therefore ask that, if you are interested in attending, to please communicate your interest through the MIAPA discussion group [6]. This will also allow us to include you in pre-workshop planning. Since the workshop is part of the TDWG conference, participants will need to register either for the full conference, or for the days of the workshop. The organizers will provide an electronic venue for participants to share ideas and develop plans in advance of the workshop. After the initial presentations, participants will self-organize into task groups. Organizers Nico Celinese, University of Florida Hilmar Lapp, NESCent Jim Leebens-Mack, University of Georgia Enrico Pontelli, New Mexico State University Arlin Stoltzfus, NIST & University of Maryland References [1] Whitacre et al. (2010). Current Best Practices for Publishing Trees Electronically. http://wiki.tdwg.org/twiki/bin/view/Phylogenetics/LinkingTrees2010 [2] O’Meara et al. (2011). Publishing re-usable phylogenetic trees, in theory and practice. Available from Nature Precedings <http://dx.doi.org/10.1038/npre.2011.6048.1> [3] Leebens-Mack, J., T. Vision, et al. (2006). "Taking the first steps towards a standard for reporting on phylogenies: Minimum Information About a Phylogenetic Analysis (MIAPA)." Omics 10(2): 231-7. [4] Whitlock, M., M. McPeek, M. Rausher, L. Rieseberg, and A. Moore (2010). Data Archiving (Editorial). The American Naturalist 175(2): 145. [5] Taylor, C.F., D. Field, S. Sansone, J. Aerts, R. Apweiler, M. Ashburner, C.A. Ball, et al. (2008). Promoting coherent minimum reporting guidelines for biological and biomedical investigations: the MIBBI project. Nature Biotechnology 26(8): 889-96. doi:10.1038/nbt.1411 [6] MIAPA discussion group: http://groups.google.com/group/miapa-discuss Published by Google Docs–Report Abuse–Updated automatically every 5 minutes |
|
From: Arlin S. <ar...@um...> - 2011-07-25 17:55:24
|
Dear all-- I would like to step down as de facto leader of the CDAO project. I have not been able to give the project the time that it deserves, nor to obtain major funding for it. I'm still interested in being involved, but I think the project would benefit from new leadership. This would be a good occasion to discuss what anyone of you expects from the project, and whether it should be re-aligned completely relative to some other initiatives. I invite everyone on this list to express their views on this issue, and to encourage this, I intend to stay quiet (at least until the discussion heats up). To ground the discussion, let me give a brief account of the history and current state of the project. CDAO grew out of a deliberate attempt by the NESCent evolutionary informatics working group (2006 to 2009) to develop an interoperability "stack" consisting of an ontology, a file format and a web services standard (phyloWS). The group developed extensive knowledge of use-cases and prior art, and documented some of this with wiki pages on use cases and concept definitions. We agreed on the importance of a "central unifying artefact" for interoperability. Some folks with a practical attitude saw the CUA as a file format, and began to work on NeXML (www.nexml.org); others saw the CUA as an ontology and began to work on CDAO. Eventually we saw these as complementary strategies. NeXML is now supported by a variety of software and is positioned to become the next file format standard in comparative analysis. From the start, the CDAO project has been the work of a small group-- with a core of just a half-dozen people-- that includes domain experts and knowledge representation experts. With help from NMSU and NESCent, we have had 3 small meetings. We developed an initial draft, worked another year to get to the "1.0" version, and published a paper describing it in 2009. After that, we started working on the next version. In practice, we have been adding stuff in creative bursts, and then going back to clean it up and re-think, rather than patiently adding one carefully documented concept at a time. We tried having regular teleconferences early in 2010 but gave up due to low participation. This is a small project. The current version of CDAO is a bit bloated and messy, in my opinion. Enrico and others have suggested splitting it into an ontology for trees, and an ontology for comparative data, knitted together in a larger ontology. CDAO does not take a clear or meaningful position on some upper-level ontology issues, which are very complex for comparative data. Once upon a time, I imagined that such representation issues would be worked out by developing critical tests based on use-cases. This hasn't happened for us-- we have benefited from some people trying to use CDAO, but mainly in terms of very low-level technical details rather than high-level concepts. With respect to applications of CDAO, there are a few. It is used in Vivek Gopolan's Nexplorer3 application, which is not yet a production application. It also plays a key role as the source of semantics in the NeXML team's metadata representation strategy. For the latter reason, it is used internally in the Phenoscape project, and it may also be used in TreeBASE IO (not sure, at present). If the CDAO project were to shut down today, then people who want to do rich metadata annotation in NeXML would have to find other sources of semantic anchors. Perhaps a way to revitalize the CDAO project would be to link development more closely to TreeBASE or Phenoscape or NeXML. Arlin ------- Arlin Stoltzfus (ar...@um...) Fellow, IBBR; Adj. Assoc. Prof., UMCP; Research Biologist, NIST IBBR, 9600 Gudelsky Drive, Rockville, MD tel: 240 314 6208; web: www.molevol.org |
|
From: Gopalan, V. (NIH/N. [C] <gop...@ni...> - 2011-07-21 17:59:33
|
Hi Hilmar, Thanks for your suggestion. You are right, It was a wrong behavior with Nexplorer. Now, I have setup Jena (Java-based ontology library I use in Nexplorer) to run "offline" by specifying alternate location for the schema urls to local files instead of fetching contents from actual namespace url. I hope it may speed-up initial loading of CDAO RDF files too. BTW, The CDAO namespace links are working fine now. Thanks again, Vivek On 7/20/11 6:37 AM, "Hilmar Lapp" <hl...@ne...> wrote: >Hi Vivek: > >While I agree that if the file in question was moved or renamed or >whatever causes the sudden 404, the error in behavior here is on >Nexplorer. Namespaces are not required to resolve in the W3C standard, >and so if your tool stops working in such a case, you should fix that. > >-hilmar > >Sent with a tap. > >On Jul 19, 2011, at 4:20 PM, "Gopalan, Vivek (NIH/NIAID) [C]" ><gop...@ni...> wrote: > >> Hi CDAO team, >> >> The base CDAO namespace >>http://www.evolutionaryontology.org/cdao/1.0/cdao.owl is not working >>since yesterday (July 18th 2011). >> >> I am getting 404 HTTP response from the redirected sourceforge URL. >> >>http://cdao.cvs.sourceforge.net/viewvc/cdao/cdao/OWL/cdao.owl?revision=1. >>34 >> >> Nexplorer3 site (http://exon.niaid.nih.gov/nexplorer) is not working >>because of this issue. Please fix this error? >> >> I verified CVS and SVN links for other sourceforge projects (nexplorer >>and cdaotools) and it seems to work fine. >> >> Thanks, >> Vivek >> >> >> >>------------------------------------------------------------------------- >>----- >> Magic Quadrant for Content-Aware Data Loss Prevention >> Research study explores the data loss prevention market. Includes >>in-depth >> analysis on the changes within the DLP market, and the criteria used to >> evaluate the strengths and weaknesses of these DLP solutions. >> http://www.accelacomm.com/jaw/sfnl/114/51385063/ >> _______________________________________________ >> CDAO-discuss mailing list >> CDA...@li... >> https://lists.sourceforge.net/lists/listinfo/cdao-discuss |
|
From: Hilmar L. <hl...@ne...> - 2011-07-20 20:30:30
|
Hi Vivek: While I agree that if the file in question was moved or renamed or whatever causes the sudden 404, the error in behavior here is on Nexplorer. Namespaces are not required to resolve in the W3C standard, and so if your tool stops working in such a case, you should fix that. -hilmar Sent with a tap. On Jul 19, 2011, at 4:20 PM, "Gopalan, Vivek (NIH/NIAID) [C]" <gop...@ni...> wrote: > Hi CDAO team, > > The base CDAO namespace http://www.evolutionaryontology.org/cdao/1.0/cdao.owl is not working since yesterday (July 18th 2011). > > I am getting 404 HTTP response from the redirected sourceforge URL. > http://cdao.cvs.sourceforge.net/viewvc/cdao/cdao/OWL/cdao.owl?revision=1.34 > > Nexplorer3 site (http://exon.niaid.nih.gov/nexplorer) is not working because of this issue. Please fix this error? > > I verified CVS and SVN links for other sourceforge projects (nexplorer and cdaotools) and it seems to work fine. > > Thanks, > Vivek > > > ------------------------------------------------------------------------------ > Magic Quadrant for Content-Aware Data Loss Prevention > Research study explores the data loss prevention market. Includes in-depth > analysis on the changes within the DLP market, and the criteria used to > evaluate the strengths and weaknesses of these DLP solutions. > http://www.accelacomm.com/jaw/sfnl/114/51385063/ > _______________________________________________ > CDAO-discuss mailing list > CDA...@li... > https://lists.sourceforge.net/lists/listinfo/cdao-discuss |
|
From: Gopalan, V. (NIH/N. [C] <gop...@ni...> - 2011-07-19 14:54:33
|
Hi CDAO team, The base CDAO namespace http://www.evolutionaryontology.org/cdao/1.0/cdao.owl is not working since yesterday (July 18th 2011). I am getting 404 HTTP response from the redirected sourceforge URL. http://cdao.cvs.sourceforge.net/viewvc/cdao/cdao/OWL/cdao.owl?revision=1.34 Nexplorer3 site (http://exon.niaid.nih.gov/nexplorer) is not working because of this issue. Please fix this error? I verified CVS and SVN links for other sourceforge projects (nexplorer and cdaotools) and it seems to work fine. Thanks, Vivek |
|
From: Hilmar L. <hl...@ne...> - 2011-04-25 17:13:06
|
Hi Arlin - your message on the 15th did make it through (it's in the archives also). And I responded to your last paragraph a day later or so. Did you receive that? -hilmar On Apr 25, 2011, at 12:46 PM, Arlin Stoltzfus wrote: > I incorporated these changes into CDAO and it is currently what you > will get here: > > http://cdao.cvs.sourceforge.net/viewvc/cdao/cdao/OWL/cdao.owl > > this is version 1.36. FYI, search the file for versionInfo or $Id > and you will find the CVS version string which is the content of > owl:versionInfo element in the XML file. Version 1.35 has the > change that Hilmar suggested to the MAO URL, and 1.36 has all the > logical changes. The revision log makes reference to this email > correspondence and to the URL with the explanation that Jim gives > below. > > Sorry about the delay. I actually did this on the 15th, right > before leaving for vacation, but an Apple mail hang-up resulted in > the loss of my message announcing the change, and I did not have > time to re-write it. > > Thanks for doing this. If you two are considering further changes, > let me know and I'll sign you up as developers. > > Arlin > > On Mar 29, 2011, at 10:16 PM, Jim Balhoff wrote: > >> Hi all, >> >> Hilmar and I have been looking at what would be required to make >> CDAO compliant with OWL 2 DL. It would be nice to be able to use >> CDAO with the standard OWL reasoners, but the current file I'm >> using (which I think is CDAO v1) includes some axioms which are not >> allowed in OWL DL. These all involve the use of non-simple >> properties in cardinality restrictions. Here is the definition of a >> simple property for OWL DL: >> >> http://www.w3.org/TR/2009/REC-owl2-syntax-20091027/#def_simple >> >> We have made some changes which we think retain all the intended >> semantics but express these concepts in a DL-compliant manner. I've >> created a table which compares the old and new class definitions: >> >> https://www.nescent.org/wg_evoinfo/CDAO_changes_for_DL-compliance >> >> I also uploaded the edited CDAO file: >> >> https://www.nescent.org/wg/evoinfo/images/5/56/Cdao-dl.owl.zip >> >> Without making some changes to be DL-compliant, importing CDAO into >> any other ontologies or datasets makes it impossible to use >> standard reasoners. If it is deemed useful to continue adding >> statements that go beyond DL, perhaps a DL version of CDAO could be >> provided, and then another file which imports that one and contains >> additional axioms could be provided. >> >> The file I uploaded works with the HermiT reasoner. However FaCT++ >> has an issue with the anySimpleType datatype. I think this may be a >> FaCT++ deficiency so I haven't yet tried to make any changes >> addressing that. >> >> Thanks, >> Jim >> ____________________________________________ >> James P. Balhoff, Ph.D. >> National Evolutionary Synthesis Center >> 2024 West Main St., Suite A200 >> Durham, NC 27705 >> USA >> >> >> >> ------------------------------------------------------------------------------ >> Enable your software for Intel(R) Active Management Technology to >> meet the >> growing manageability and security demands of your customers. >> Businesses >> are taking advantage of Intel(R) vPro (TM) technology - will your >> software >> be a part of the solution? Download the Intel(R) Manageability >> Checker >> today! http://p.sf.net/sfu/intel-dev2devmar >> _______________________________________________ >> CDAO-discuss mailing list >> CDA...@li... >> https://lists.sourceforge.net/lists/listinfo/cdao-discuss > > ------- > Arlin Stoltzfus (ar...@um...) > Fellow, IBBR; Adj. Assoc. Prof., UMCP; Research Biologist, NIST > IBBR, 9600 Gudelsky Drive, Rockville, MD > tel: 240 314 6208; web: www.molevol.org > -- =========================================================== : Hilmar Lapp -:- Durham, NC -:- informatics.nescent.org : =========================================================== |
|
From: Arlin S. <ar...@um...> - 2011-04-25 16:47:03
|
I incorporated these changes into CDAO and it is currently what you will get here: http://cdao.cvs.sourceforge.net/viewvc/cdao/cdao/OWL/cdao.owl this is version 1.36. FYI, search the file for versionInfo or $Id and you will find the CVS version string which is the content of owl:versionInfo element in the XML file. Version 1.35 has the change that Hilmar suggested to the MAO URL, and 1.36 has all the logical changes. The revision log makes reference to this email correspondence and to the URL with the explanation that Jim gives below. Sorry about the delay. I actually did this on the 15th, right before leaving for vacation, but an Apple mail hang-up resulted in the loss of my message announcing the change, and I did not have time to re- write it. Thanks for doing this. If you two are considering further changes, let me know and I'll sign you up as developers. Arlin On Mar 29, 2011, at 10:16 PM, Jim Balhoff wrote: > Hi all, > > Hilmar and I have been looking at what would be required to make > CDAO compliant with OWL 2 DL. It would be nice to be able to use > CDAO with the standard OWL reasoners, but the current file I'm using > (which I think is CDAO v1) includes some axioms which are not > allowed in OWL DL. These all involve the use of non-simple > properties in cardinality restrictions. Here is the definition of a > simple property for OWL DL: > > http://www.w3.org/TR/2009/REC-owl2-syntax-20091027/#def_simple > > We have made some changes which we think retain all the intended > semantics but express these concepts in a DL-compliant manner. I've > created a table which compares the old and new class definitions: > > https://www.nescent.org/wg_evoinfo/CDAO_changes_for_DL-compliance > > I also uploaded the edited CDAO file: > > https://www.nescent.org/wg/evoinfo/images/5/56/Cdao-dl.owl.zip > > Without making some changes to be DL-compliant, importing CDAO into > any other ontologies or datasets makes it impossible to use standard > reasoners. If it is deemed useful to continue adding statements that > go beyond DL, perhaps a DL version of CDAO could be provided, and > then another file which imports that one and contains additional > axioms could be provided. > > The file I uploaded works with the HermiT reasoner. However FaCT++ > has an issue with the anySimpleType datatype. I think this may be a > FaCT++ deficiency so I haven't yet tried to make any changes > addressing that. > > Thanks, > Jim > ____________________________________________ > James P. Balhoff, Ph.D. > National Evolutionary Synthesis Center > 2024 West Main St., Suite A200 > Durham, NC 27705 > USA > > > > ------------------------------------------------------------------------------ > Enable your software for Intel(R) Active Management Technology to > meet the > growing manageability and security demands of your customers. > Businesses > are taking advantage of Intel(R) vPro (TM) technology - will your > software > be a part of the solution? Download the Intel(R) Manageability Checker > today! http://p.sf.net/sfu/intel-dev2devmar > _______________________________________________ > CDAO-discuss mailing list > CDA...@li... > https://lists.sourceforge.net/lists/listinfo/cdao-discuss ------- Arlin Stoltzfus (ar...@um...) Fellow, IBBR; Adj. Assoc. Prof., UMCP; Research Biologist, NIST IBBR, 9600 Gudelsky Drive, Rockville, MD tel: 240 314 6208; web: www.molevol.org |
|
From: Hilmar L. <hl...@ne...> - 2011-04-16 03:14:16
|
One of the things that I'd like to do is modularize the ontology a bit further. For example, my understanding so far is that the relationships to MAO terms augment MAO terms (and by reciprocity CDAO terms, too, of course), but they aren't necessary to define the semantics of CDAO terms. I.e., the CDAO terms can live on their own, without MAO imported and without the relationships to MAO terms, while having unaltered semantics. If that is true, it'd be useful to put the MAO import and CDAO-MAO relationships into a separate file (ontology) that imports both CDAO and MAO, and then adds the relationships. Similarly with the co-ode amino acid ontology - is this really needed in this ontology? It brings with it a lot of terms (and it is imported in full) and properties, most of which aren't used. Modularizing the files can help in making reasoning faster for applications that use CDAO but don't require those imports. -hilmar On Apr 15, 2011, at 8:56 AM, Arlin Stoltzfus wrote: > OK, I have now looked at all the changes in detail. Does anyone > have tricks for diffing these files? I had to run them both through > the same version of Protege, and then sort through the spurious > diffs due to different orderings of statements. > > The file you were using was 1.33, not the "1.0" version which is > 1.34. But that doesn't matter much because the only difference was > the MAO URL, which Hilmar says needs to be updated anyway. > > So, I committed these changes. 1.35 has the DL-compliance changes, > and 1.36 has the MAO URL fix. You can get it here: > > http://cdao.cvs.sourceforge.net/viewvc/cdao/cdao/OWL/cdao.owl?revision=1.36 > > If no problems emerge from this in the next few weeks, I'll change > the URL mapping so that this version corresponds to the "1.0" > namespace (i.e., evolutionaryontology.org/cdao/1.0/cdao.owl). > > Thanks for doing this! If you guys are going to do more, I would be > happy to put you on the access list as developers. > > Arlin > > On Mar 30, 2011, at 12:03 AM, Hilmar Lapp wrote: > >> BTW interestingly I could classify this using Fact++ in Protege >> 4.0.2. >> Now that I've upgraded to 4.1 classifying with Fact++ crashes >> Protege. >> HermiT does work. >> >> BTW the MAO import URL needs to be changed to http://purl.org/obo/obo-all/mao/mao.owl >> . I don't know when this was changed and why (and it's not compliant >> with the new ID schema anyway, so I suspect complaining won't get us >> much). The file also contains an import of an ontology in Enrico's >> webserver home directory - is that necessary? >> >> Note also that the "belongs_to_Edge_as_Parent exactly 2 owl:Thing" >> from "has_Child exactly 2 Node" isn't losing the typing of the >> cardinality restriction, because belongs_to_Edge_as_Parent has range >> DirectedEdge. >> >> As Jim says, if this seems too difficult to merge these changes into >> the main ontology, I or Jim would be happy to commit this version to >> cvs. I do need a classifiable version for several purposes. >> >> -hilmar >> >> On Mar 29, 2011, at 10:16 PM, Jim Balhoff wrote: >> >>> Hi all, >>> >>> Hilmar and I have been looking at what would be required to make >>> CDAO compliant with OWL 2 DL. It would be nice to be able to use >>> CDAO with the standard OWL reasoners, but the current file I'm using >>> (which I think is CDAO v1) includes some axioms which are not >>> allowed in OWL DL. These all involve the use of non-simple >>> properties in cardinality restrictions. Here is the definition of a >>> simple property for OWL DL: >>> >>> http://www.w3.org/TR/2009/REC-owl2-syntax-20091027/#def_simple >>> >>> We have made some changes which we think retain all the intended >>> semantics but express these concepts in a DL-compliant manner. I've >>> created a table which compares the old and new class definitions: >>> >>> https://www.nescent.org/wg_evoinfo/CDAO_changes_for_DL-compliance >>> >>> I also uploaded the edited CDAO file: >>> >>> https://www.nescent.org/wg/evoinfo/images/5/56/Cdao-dl.owl.zip >>> >>> Without making some changes to be DL-compliant, importing CDAO into >>> any other ontologies or datasets makes it impossible to use standard >>> reasoners. If it is deemed useful to continue adding statements that >>> go beyond DL, perhaps a DL version of CDAO could be provided, and >>> then another file which imports that one and contains additional >>> axioms could be provided. >>> >>> The file I uploaded works with the HermiT reasoner. However FaCT++ >>> has an issue with the anySimpleType datatype. I think this may be a >>> FaCT++ deficiency so I haven't yet tried to make any changes >>> addressing that. >>> >>> Thanks, >>> Jim >>> ____________________________________________ >>> James P. Balhoff, Ph.D. >>> National Evolutionary Synthesis Center >>> 2024 West Main St., Suite A200 >>> Durham, NC 27705 >>> USA >>> >>> >> >> -- >> =========================================================== >> : Hilmar Lapp -:- Durham, NC -:- informatics.nescent.org : >> =========================================================== >> >> >> >> >> ------------------------------------------------------------------------------ >> Enable your software for Intel(R) Active Management Technology to >> meet the >> growing manageability and security demands of your customers. >> Businesses >> are taking advantage of Intel(R) vPro (TM) technology - will your >> software >> be a part of the solution? Download the Intel(R) Manageability >> Checker >> today! http://p.sf.net/sfu/intel-dev2devmar >> _______________________________________________ >> CDAO-discuss mailing list >> CDA...@li... >> https://lists.sourceforge.net/lists/listinfo/cdao-discuss > > ------- > Arlin Stoltzfus (ar...@um...) > Fellow, IBBR; Adj. Assoc. Prof., UMCP; Research Biologist, NIST > IBBR, 9600 Gudelsky Drive, Rockville, MD > tel: 240 314 6208; web: www.molevol.org > -- =========================================================== : Hilmar Lapp -:- Durham, NC -:- informatics.nescent.org : =========================================================== |
|
From: Arlin S. <ar...@um...> - 2011-04-15 12:56:29
|
OK, I have now looked at all the changes in detail. Does anyone have tricks for diffing these files? I had to run them both through the same version of Protege, and then sort through the spurious diffs due to different orderings of statements. The file you were using was 1.33, not the "1.0" version which is 1.34. But that doesn't matter much because the only difference was the MAO URL, which Hilmar says needs to be updated anyway. So, I committed these changes. 1.35 has the DL-compliance changes, and 1.36 has the MAO URL fix. You can get it here: http://cdao.cvs.sourceforge.net/viewvc/cdao/cdao/OWL/cdao.owl?revision=1.36 If no problems emerge from this in the next few weeks, I'll change the URL mapping so that this version corresponds to the "1.0" namespace (i.e., evolutionaryontology.org/cdao/1.0/cdao.owl). Thanks for doing this! If you guys are going to do more, I would be happy to put you on the access list as developers. Arlin On Mar 30, 2011, at 12:03 AM, Hilmar Lapp wrote: > BTW interestingly I could classify this using Fact++ in Protege 4.0.2. > Now that I've upgraded to 4.1 classifying with Fact++ crashes Protege. > HermiT does work. > > BTW the MAO import URL needs to be changed to http://purl.org/obo/obo-all/mao/mao.owl > . I don't know when this was changed and why (and it's not compliant > with the new ID schema anyway, so I suspect complaining won't get us > much). The file also contains an import of an ontology in Enrico's > webserver home directory - is that necessary? > > Note also that the "belongs_to_Edge_as_Parent exactly 2 owl:Thing" > from "has_Child exactly 2 Node" isn't losing the typing of the > cardinality restriction, because belongs_to_Edge_as_Parent has range > DirectedEdge. > > As Jim says, if this seems too difficult to merge these changes into > the main ontology, I or Jim would be happy to commit this version to > cvs. I do need a classifiable version for several purposes. > > -hilmar > > On Mar 29, 2011, at 10:16 PM, Jim Balhoff wrote: > >> Hi all, >> >> Hilmar and I have been looking at what would be required to make >> CDAO compliant with OWL 2 DL. It would be nice to be able to use >> CDAO with the standard OWL reasoners, but the current file I'm using >> (which I think is CDAO v1) includes some axioms which are not >> allowed in OWL DL. These all involve the use of non-simple >> properties in cardinality restrictions. Here is the definition of a >> simple property for OWL DL: >> >> http://www.w3.org/TR/2009/REC-owl2-syntax-20091027/#def_simple >> >> We have made some changes which we think retain all the intended >> semantics but express these concepts in a DL-compliant manner. I've >> created a table which compares the old and new class definitions: >> >> https://www.nescent.org/wg_evoinfo/CDAO_changes_for_DL-compliance >> >> I also uploaded the edited CDAO file: >> >> https://www.nescent.org/wg/evoinfo/images/5/56/Cdao-dl.owl.zip >> >> Without making some changes to be DL-compliant, importing CDAO into >> any other ontologies or datasets makes it impossible to use standard >> reasoners. If it is deemed useful to continue adding statements that >> go beyond DL, perhaps a DL version of CDAO could be provided, and >> then another file which imports that one and contains additional >> axioms could be provided. >> >> The file I uploaded works with the HermiT reasoner. However FaCT++ >> has an issue with the anySimpleType datatype. I think this may be a >> FaCT++ deficiency so I haven't yet tried to make any changes >> addressing that. >> >> Thanks, >> Jim >> ____________________________________________ >> James P. Balhoff, Ph.D. >> National Evolutionary Synthesis Center >> 2024 West Main St., Suite A200 >> Durham, NC 27705 >> USA >> >> > > -- > =========================================================== > : Hilmar Lapp -:- Durham, NC -:- informatics.nescent.org : > =========================================================== > > > > > ------------------------------------------------------------------------------ > Enable your software for Intel(R) Active Management Technology to > meet the > growing manageability and security demands of your customers. > Businesses > are taking advantage of Intel(R) vPro (TM) technology - will your > software > be a part of the solution? Download the Intel(R) Manageability Checker > today! http://p.sf.net/sfu/intel-dev2devmar > _______________________________________________ > CDAO-discuss mailing list > CDA...@li... > https://lists.sourceforge.net/lists/listinfo/cdao-discuss ------- Arlin Stoltzfus (ar...@um...) Fellow, IBBR; Adj. Assoc. Prof., UMCP; Research Biologist, NIST IBBR, 9600 Gudelsky Drive, Rockville, MD tel: 240 314 6208; web: www.molevol.org |
|
From: Hilmar L. <hl...@ne...> - 2011-04-07 22:36:12
|
On Apr 7, 2011, at 1:04 PM, Jim Balhoff wrote: > part_of isn't defined as transitive in that copy of CDAO Ah - I know why. It is imported from MAO, and in my copy I changed the import to actually import something. (OBO files cannot be imported into an OWL ontology. But there is an OWL version MAO.) If I make part_of non-transitive, Fact++ still crashes though for me, so that's not the sticking point. -hilmar -- =========================================================== : Hilmar Lapp -:- Durham, NC -:- informatics.nescent.org : =========================================================== |