Menu
▾
▴
bodington-developers — Mailing list for Bodington developers
You can subscribe to this list here.
| 2005 |
Jan
|
Feb
|
Mar
|
Apr
(41) |
May
(353) |
Jun
(133) |
Jul
(534) |
Aug
(401) |
Sep
(219) |
Oct
(86) |
Nov
(144) |
Dec
(61) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2006 |
Jan
(200) |
Feb
(130) |
Mar
(345) |
Apr
(153) |
May
(247) |
Jun
(338) |
Jul
(222) |
Aug
(70) |
Sep
(39) |
Oct
(27) |
Nov
(76) |
Dec
(30) |
| 2007 |
Jan
(81) |
Feb
(44) |
Mar
(9) |
Apr
|
May
(3) |
Jun
(2) |
Jul
(34) |
Aug
(2) |
Sep
(1) |
Oct
|
Nov
|
Dec
(6) |
| 2008 |
Jan
|
Feb
|
Mar
(5) |
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| 2009 |
Jan
(1) |
Feb
|
Mar
|
Apr
|
May
(7) |
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: Jon M. <jo...@te...> - 2006-03-01 14:18:17
|
I won't hear anything said against MS-SQL. I think it's one of the few MS products that I would recommend - mainly for the profiler, query analyser and management user interface. Only problem is the silly price. Jon Peter Crowther wrote: >>From: Jon Maber >>I havn't done anything with oracle for a long >>time so it can't have been a recent version. >> >> > >Lucky. There are a lot of reformed Oracle users out there - it's a bit >like reformed smokers ;-). Ex-MSSQL users could no doubt be named as >recovering from some similar addiction! > > - Peter > > >------------------------------------------------------- >This SF.Net email is sponsored by xPML, a groundbreaking scripting language >that extends applications into web and mobile media. Attend the live webcast >and join the prime developer group breaking into this new coding territory! >http://sel.as-us.falkag.net/sel?cmd=k&kid0944&bid$1720&dat1642 >_______________________________________________ >Bodington-developers mailing list >Bod...@li... >https://lists.sourceforge.net/lists/listinfo/bodington-developers > > > |
|
From: Peter C. <Pet...@me...> - 2006-03-01 14:04:57
|
> From: Jon Maber > I havn't done anything with oracle for a long=20 > time so it can't have been a recent version. Lucky. There are a lot of reformed Oracle users out there - it's a bit like reformed smokers ;-). Ex-MSSQL users could no doubt be named as recovering from some similar addiction! - Peter |
|
From: Jon M. <jo...@te...> - 2006-03-01 14:02:10
|
Matthew Buckett wrote: >Peter Crowther wrote: > > >>>From: Jon Maber >>>However, some database products (oracle?) think "" == null >>> >>> > >Yeah I seem to remember reading a comment about that. Do you know which >version had the problem? > > No, sorry - but I havn't done anything with oracle for a long time so it can't have been a recent version. > > >>SQL Server pre-7.0 had that particular delusion. I seen to recall it's >>mandated by one of the SQL standards, in fact. >> >> >> >> >>>Could we put "unknown" or "unset" in instead of null? >>> >>> >>I'd prefer to alter the schema if we're going to have that kind of >>special value - personal opinion, I freely accept. >> >> > >I prefer the null as there are less differences between the upper APIs >and the lower ones so developers don't get confused. I reaslise this is >why you have abstraction but it helps if you don't have to use it when >things are white boxes. > > > You're probably right. Jon |
|
From: Alexis O'C. <ale...@co...> - 2006-03-01 14:01:59
|
Peter Crowther wrote: >> From: Adam Marshall >> I take it you have pointed Mike in the direction of Alexis's's's >> questionnaire modifications. > > Yes. The issue is not with the display; the issue is with the business > rule for when the tutor should be allowed to see the list. > > - Peter > Cough, cough, ahem. Yes, currently the "business rule" in question says 'yes' when the number of respondents is greater than 2. We haven't fully decided on what the criteria should be at Oxford. I considered making it more of a 'pluggable point', i.e. interface, so maybe the algorithm could operate on the basis of specified '% responded', 'total no. of respondents' via a configuration file(?). Hmmm ... Alexis |
|
From: Jon M. <jo...@te...> - 2006-03-01 13:59:13
|
For accessibility I'd suggest you follow the pattern set in messaging rooms. The user can choose how to navigate and you have all the CSS styles you need for formatting in 'n' different ways according to user preference. >Depending on how much JavaScript is acceptable, I've got a rather nice >tree system that I developed for another project that can query for a >list of children of a node and modify the tree on-the-fly. I've no >doubt there are better-developed and maintained third-party systems that >do it as well, likely using AJAX or similar. > >Plays merry hell with accessibility though. > > - Peter > > |
|
From: Alexis O'C. <ale...@co...> - 2006-03-01 13:56:15
|
Andrew Booth wrote: > We moved from NetBeans and JBuilder to Eclipse a while ago. > The Sun format doesn't seem to be one of the formats in Eclipse by default. > Any idea how we can load it without having to define it from scratch? > > Aggie > It's actually the one called 'Java Conventions'. The Eclipse team realized that they weren't actually following the Sun guidelines to the letter, so that's why they renamed their own ones(s) as 'Eclipse' & 'Eclipse v2.1'(!) There are further considerations such as strictly speaking Sun says that even single line statements should be within curly braces, which is something I personally wholeheartedly endorse, but means more diffs. Another possibility is to use of the 2 distinct 'Jon Maber' styles that I identified and have available as Eclipse code formatter templates ;-) ... Alexis |
|
From: Matthew B. <mat...@ou...> - 2006-03-01 13:48:07
|
Peter Crowther wrote: >>From: Jon Maber >>However, some database products (oracle?) think "" == null Yeah I seem to remember reading a comment about that. Do you know which version had the problem? > SQL Server pre-7.0 had that particular delusion. I seen to recall it's > mandated by one of the SQL standards, in fact. > > >>Could we put "unknown" or "unset" in instead of null? > > > I'd prefer to alter the schema if we're going to have that kind of > special value - personal opinion, I freely accept. I prefer the null as there are less differences between the upper APIs and the lower ones so developers don't get confused. I reaslise this is why you have abstraction but it helps if you don't have to use it when things are white boxes. -- -- Matthew Buckett, VLE Developer -- Learning Technologies Group, Oxford University Computing Services -- Tel: +44 (0)1865 283660 http://www.oucs.ox.ac.uk/ltg/ |
|
From: Peter C. <Pet...@me...> - 2006-03-01 13:45:31
|
> From: Matthew Buckett > This is one problem we have with Bodington currently not having a > concept of roles. Rather than just piling on the permissions are we > going to ever look defining and using roles? Quite. > Do you have anything more on what is considered editing the=20 > resource? No, but I'm gathering that information. - Peter |
|
From: Andrew B. <a.g...@le...> - 2006-03-01 13:45:07
|
We moved from NetBeans and JBuilder to Eclipse a while ago.
The Sun format doesn't seem to be one of the formats in Eclipse by default.
Any idea how we can load it without having to define it from scratch?
Aggie
-----Original Message-----
From: bod...@li...
[mailto:bod...@li...] On Behalf Of
Matthew Buckett
Sent: 01 March 2006 13:34
To: bod...@li...
Subject: Re: [Bodington-developers] Code Formatting
Peter Crowther wrote:
>>From: Jon Maber
>>Perhaps, a standard format
>>based on common capabilities of the main editors could be agreed.
>
>
> Could it be the Sun one? It's the one most likely to be supported by
> all the editors, and it's the one most widely used on other OS Java
> projects.
Would get my support, although it means the { get moved onto a different
like but I think it is worth it as it is the one most people are used to
reading and is already heavily documented.
--
-- Matthew Buckett, VLE Developer
-- Learning Technologies Group, Oxford University Computing Services
-- Tel: +44 (0)1865 283660 http://www.oucs.ox.ac.uk/ltg/
-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
_______________________________________________
Bodington-developers mailing list
Bod...@li...
https://lists.sourceforge.net/lists/listinfo/bodington-developers
|
|
From: Matthew B. <mat...@ou...> - 2006-03-01 13:41:46
|
Peter Crowther wrote: > Access right addition > ===================== > > Dramatically simplify granting access for QA or external examiner > reviews. > > a) Addition of a new access right for reviewers. This will work > identically to Administer, but will only provide cascaded See and View > access. This will be called "Evaluate" This is one problem we have with Bodington currently not having a concept of roles. Rather than just piling on the permissions are we going to ever look defining and using roles? > View Access addition > ==================== > > Simplify resource/teacher management for Floor Managers and others. > > Addition to the Visual Access Rights tool, that displays information > about the owners and editors. This will: > a) list the names and usernames of those people in the owners group On WebLearn we display the owners of the resource menu page at the moment. > b) display the name and username of the person who created the > resource, along with the date of creation > > c) display the name and username of the last person to edit the > resource as well as the date/time of the edit. Do you have anything more on what is considered editing the resource? Is adding a child to a container considered changing it? Is uploading a file to a resource? Or it is just editing the basic resource properties (title, introduction, description). -- -- Matthew Buckett, VLE Developer -- Learning Technologies Group, Oxford University Computing Services -- Tel: +44 (0)1865 283660 http://www.oucs.ox.ac.uk/ltg/ |
|
From: Peter C. <Pet...@me...> - 2006-03-01 13:36:22
|
> From: Matthew Buckett > Thats what I though although I don't have any experience of=20 > joining this > many tables with reasonable numbers of rows. All of the tables are in > the 10'000s here at Oxford. It does rather depend on good indexing - and, in some cases, how good the RDBMS is at noticing or being forced to use covering indexes. SQL Server's always been pretty good at this, even pre-7.0; the OS systems vary widely. Covering indexes for the purpose of the query give you a massive boost, as you don't have to retrieve the rest of the data on the page and they're automatically sorted in the correct order for the query. If you assume that you'll also get good selectivity out of the resource id (a reasonable assumption as most resources will have at most low-tens of ACEs) and the user to group mapping (users will typically be in a few tens of groups) then the RDBMS can optimise pretty well. - Peter |
|
From: Matthew B. <mat...@ou...> - 2006-03-01 13:33:55
|
Peter Crowther wrote:
>>From: Jon Maber
>>Perhaps, a standard format
>>based on common capabilities of the main editors could be agreed.
>
>
> Could it be the Sun one? It's the one most likely to be supported by
> all the editors, and it's the one most widely used on other OS Java
> projects.
Would get my support, although it means the { get moved onto a different
like but I think it is worth it as it is the one most people are used to
reading and is already heavily documented.
--
-- Matthew Buckett, VLE Developer
-- Learning Technologies Group, Oxford University Computing Services
-- Tel: +44 (0)1865 283660 http://www.oucs.ox.ac.uk/ltg/
|
|
From: Peter C. <Pet...@me...> - 2006-03-01 13:33:24
|
> From: Matthew Buckett > There are some formatters that can be called from ant and=20 > using that as > the primary formatter means that anyone who can compile Bodington can > just run the formatter over the code before committing changes. Nice. Like it. Anything to stop me having to think about it :-). - Peter |
|
From: Matthew B. <mat...@ou...> - 2006-03-01 13:31:10
|
Jon Maber wrote: > > I use netbeans as a development tool and that has a reformatting tool. > Do other editors have the same kind of thing? Perhaps, a standard format > based on common capabilities of the main editors could be agreed. Yeah, Eclipse has a nice formatter. > I would suggest that systematic reformatting of the source files be > delayed until the next major release. This will avoiding muddying diffs > of files versions within the current development cycle. How about this; > > 1) Agree now a standard format and describe the settings required for > main editors to auto format. There are some formatters that can be called from ant and using that as the primary formatter means that anyone who can compile Bodington can just run the formatter over the code before committing changes. Eg: http://sourceforge.net/projects/jalopy/ > 2) Use that for all new classes. > 3) Complete development cycle up to release which is tagged. > 4) Immediately implement a code tidying cycle and tag that version too. Seems the way to go. > the next major version is labelled in the CVS -- -- Matthew Buckett, VLE Developer -- Learning Technologies Group, Oxford University Computing Services -- Tel: +44 (0)1865 283660 http://www.oucs.ox.ac.uk/ltg/ |
|
From: Peter C. <Pet...@me...> - 2006-03-01 13:25:55
|
> From: Jon Maber > However, some database products (oracle?) think "" =3D=3D null SQL Server pre-7.0 had that particular delusion. I seen to recall it's mandated by one of the SQL standards, in fact. > Could we put "unknown" or "unset" in instead of null? I'd prefer to alter the schema if we're going to have that kind of special value - personal opinion, I freely accept. - Peter |
|
From: Matthew B. <mat...@ou...> - 2006-03-01 13:23:45
|
Peter Crowther wrote: >>From: Matthew Buckett >>I know permission checks are caches for a little while (~5 seconds) so >>once permission have been calculated for a request they should stick >>around for a while. Resource.storeRecentPermission() is used. > > > Ah! Thanks. > > >>With the recursive >>nature of the permissions (inherit) I don't see an easy way >>todo this in >>the database without store procedures (please don't go there). > > > Awww, can I? Please? Why not, lets just use Oracle and then we can write our stored procedures in Java and everything will be great ;-) >>select resource.id from users, groups, resources, acls, >>acl_entrys where >>[join all tables together] and acl_entry.manage = true; > > > Assuming manage is broken out into its own column, is positive... > > >>Do we want permissions to be calculatable in the database? Would the >>previous SQL ever run acceptably? > > > With appropriate indexing, yes it would - at least on the RDBMSs I use. Thats what I though although I don't have any experience of joining this many tables with reasonable numbers of rows. All of the tables are in the 10'000s here at Oxford. >>We could fake the "inherit access" behind the scences >>although it makes >>some permission changes expensive. > > > Yes. But, as you pointed out, permissions change rarely; large > recursive changes are even more rare. Paying the time to update a cache > table when they change is, to my mind, a worthwhile tradeoff. It's > similar to the way NTFS stores permissions (inherited ACEs are added to > each ACL) rather than the old NetWare approach that maintained a central > list and calculated on each access. With traditional UNIX you of course don't have "inherit" but your just get chmod -R. Although this is a sligly different model I personally perfer it as you don't delegate as much control to containing resource owners, but I'm not saying we should change it for Bodington. > If you assume that space is cheaper > than time and that reads are much more common than writes, you can make > very large savings. I think it's the way to go. -- -- Matthew Buckett, VLE Developer -- Learning Technologies Group, Oxford University Computing Services -- Tel: +44 (0)1865 283660 http://www.oucs.ox.ac.uk/ltg/ |
|
From: Peter C. <Pet...@me...> - 2006-03-01 13:23:41
|
> From: Adam Marshall > I take it you have pointed Mike in the direction of Alexis's's's > questionnaire modifications. Yes. The issue is not with the display; the issue is with the business rule for when the tutor should be allowed to see the list. - Peter |
|
From: Peter C. <Pet...@me...> - 2006-03-01 13:22:43
|
> From: Jon Maber > Quite - even Windows explorer doesn't explode all the=20 > subdirectories at once when you open it. Luckily. > In essence the same kind of approach=20 > is required=20 > - checking access rights to a list of siblings and a when the=20 > user opens=20 > up another branch another list of siblings is checked. Depending on how much JavaScript is acceptable, I've got a rather nice tree system that I developed for another project that can query for a list of children of a node and modify the tree on-the-fly. I've no doubt there are better-developed and maintained third-party systems that do it as well, likely using AJAX or similar. Plays merry hell with accessibility though. - Peter |
|
From: Adam M. <ada...@co...> - 2006-03-01 13:20:24
|
| Questionnaire development | ========================= | | Provide a mechanism to monitor completion of module feedback | questionnaires without infringing anonymous submission. | | a) Allow a deadline to be set after which the owner can view a | standard HTML list of all students that have record access to the | questionnaire who have not completed it. This list will not be displayed | if less than 50% of the students who have "record" access have not | completed it. | | [This is currently under revision; expect to see an email on this list | from Michael Thomas] | | I take it you have pointed Mike in the direction of Alexis's's's questionnaire modifications. adam |
|
From: Peter C. <Pet...@me...> - 2006-03-01 13:19:54
|
> From: Adam Marshall > Would your ASK 'rewrite' be able to be used by Bodington? The concepts would; the code would not. The approach is well-known. - Peter |
|
From: Jon M. <jo...@te...> - 2006-03-01 13:18:43
|
Matthew Buckett wrote: >Adam Marshall wrote: > > >>I think it would be an absolute winner if we could have a windows explorer >>type interface showing only the resources that one can see. This would >>remove many of the gripes that people have about navigation. >> >> > >Not being able to get "all the resources which I have manage rights" >doesn't preclude the ability to have an explorer style interface as you >wouldn't want to send all the resources to the client at once (too big). > > > Quite - even Windows explorer doesn't explode all the subdirectories at once when you open it. In essence the same kind of approach is required - checking access rights to a list of siblings and a when the user opens up another branch another list of siblings is checked. On the other hand implementing a more efficient caching system that uses the database sounds like an interesting idea. If the caching lasts any significant length of time it would be necessary to invalidate chunks of it when permission changes are made. |
|
From: Peter C. <Pet...@me...> - 2006-03-01 13:16:58
|
> From: Matthew Buckett > But as you say it does impact on the history. It doesn't preclude > checking changes, it just means that if you want to look over the > reformat you need to checkout the old version, format it, then do the > diff which means only in extreme instances would most people bother. I had a look at this a while back. There are diffs out there that can be told to ignore leading whitespace, which would be the biggest single issue with a reformat to Sun standards at present. I'd have to go hunting to find them again, though. - Peter |
|
From: Peter C. <Pet...@me...> - 2006-03-01 13:14:42
|
> From: Matthew Buckett > Does > anyone else have a Bodington tree? Leeds? UHI do I think. UHI have stated in the past that they're developing on a branch at SF - or at least I seem to recall Sean saying that. - Peter |
|
From: Jon M. <jo...@te...> - 2006-03-01 13:11:25
|
I don't think there's a problem with leaving the schema as it is. However, some database products (oracle?) think "" == null which would be a problem. I.e. trying to put "" into the field will through a "nulls not allowed" error. Could we put "unknown" or "unset" in instead of null? Adam Marshall wrote: >RDBMS-specific scripts - there's already a precedent for this isn't there? > >Ada. > > > |
|
From: Peter C. <Pet...@me...> - 2006-03-01 13:10:59
|
> From: Jon Maber > I use netbeans as a development tool and that has a=20 > reformatting tool.=20 > Do other editors have the same kind of thing? All the Java IDEs do, to my knowledge. The amount of tailoring allowed varies somewhat. > Perhaps, a standard format=20 > based on common capabilities of the main editors could be agreed. Could it be the Sun one? It's the one most likely to be supported by all the editors, and it's the one most widely used on other OS Java projects. (runs screaming as a horde of angry Bodington developers chase him over the horizon) - Peter |
19 messages has been excluded from this view by a project administrator.
