servicesng-chat — General discussion about servicesng

Re: [SNG-chat] Let's start something up (was: Re: [epona] Epona 'is this the end')

[David R. <dav...@vi...>]

I wouldn't mind working on bugs / new features (if its organised we might
actaully see the bug reports so we know where they are!), but I agree
someone needs to be running it, if not, we'll all produce lots of fix's,
and they will never be incorperated :/

If anyone can contact lara, asking her to appoint a successor to carry on
development would be nice :)

I have no problems with anyone doing it long as it isnt me, i couldnt
organise anything :)

Rob


> I also disagree with portability on the services. It has
> to run on unix, that's it. If it compiles under cygwin
> great if not... most wont care.
>
> I'm sure that we, as a group can indeed improve Epona
> as we know it. Again, release bug fixes first, then
> small features, then big features. Most of us already
> have a few patches to put it. We'd need to get organized
> and do it.
>
> About services-ng, that could also happen, but I know that
> I'll give less of my time to it since it will be java based.
>
> The first thing is that somebody (or more than one) has
> to step up as the new Epona project leader to coordinate
> all our efforts. I know of a few in here that would do a
> great job at it. I have no problem being that person, I
> could easily set everything up and start getting patches
> in about a week.
>
> should we do that?
>
> d.
>
> On Thu, Jan 23, 2003 at 09:23:55PM +0200, Noam M. wrote:
>> Daniele Nicolucci (Jollino) wrote:
>> <snip>
>> >- Installing a recent JRE to make services live in a fast JVM would
>> probably be too tricky, if not impossible, in some configurations -
>> Installing gcc 3.x to compile Java stuff into native machine code
>> would probably be even worse
>> >
>> >This is because many machines running irc services are often used for
>>  other stuff, and they just can't afford the risk of upgrading the
>> JRE  (they could be using java for some stuff that can't be
>> interrupted), or  even worse, upgrading to gcc 3.x to compile
>> services. Upgrading gcc  could even mean, in the most Armageddon-like
>> situation, that the whole  system ends up being messed up by
>> accidental overwriting of libraries.  It's a possibility, although
>> it's improbable. But it's still possible.
>> >
>> I have a suggestion on how to solve this problem.
>> provide compiled versions of your services on all popular operating
>> systems. this might be tricky due to the so many operating systems out
>> there, but  I'm sure you can find ppl who are willing to compile it
>> for you. Im no expert with *nix in general, and never even seen/used
>> java as a  language, so my suggestion might not even be possible.
>>
>> -Noam M.
>>
>>
>>
>> -------------------------------------------------------
>> This SF.NET email is sponsored by:
>> SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
>> http://www.vasoftware.com
>> _______________________________________________
>> Servicesng-chat mailing list
>> Ser...@li...
>> https://lists.sourceforge.net/lists/listinfo/servicesng-chat
>
>
> -------------------------------------------------------
> This SF.NET email is sponsored by:
> SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
> http://www.vasoftware.com
> _______________________________________________
> Servicesng-chat mailing list
> Ser...@li...
> https://lists.sourceforge.net/lists/listinfo/servicesng-chat

Re: [SNG-chat] Let's start something up (was: Re: [epona] Epona "is this the end")

[Daniel E. <da...@ze...>]

I also disagree with portability on the services. It has
to run on unix, that's it. If it compiles under cygwin
great if not... most wont care.  

I'm sure that we, as a group can indeed improve Epona 
as we know it. Again, release bug fixes first, then 
small features, then big features. Most of us already 
have a few patches to put it. We'd need to get organized
and do it. 

About services-ng, that could also happen, but I know that
I'll give less of my time to it since it will be java based.

The first thing is that somebody (or more than one) has
to step up as the new Epona project leader to coordinate
all our efforts. I know of a few in here that would do a
great job at it. I have no problem being that person, I
could easily set everything up and start getting patches
in about a week. 

should we do that?

d.

On Thu, Jan 23, 2003 at 09:23:55PM +0200, Noam M. wrote:
> Daniele Nicolucci (Jollino) wrote:
> <snip>
> >- Installing a recent JRE to make services live in a fast JVM would 
> >probably be too tricky, if not impossible, in some configurations
> >- Installing gcc 3.x to compile Java stuff into native machine code 
> >would probably be even worse
> >
> >This is because many machines running irc services are often used for 
> >other stuff, and they just can't afford the risk of upgrading the JRE 
> >(they could be using java for some stuff that can't be interrupted), or 
> >even worse, upgrading to gcc 3.x to compile services. Upgrading gcc 
> >could even mean, in the most Armageddon-like situation, that the whole 
> >system ends up being messed up by accidental overwriting of libraries. 
> >It's a possibility, although it's improbable. But it's still possible.
> >
> I have a suggestion on how to solve this problem.
> provide compiled versions of your services on all popular operating systems.
> this might be tricky due to the so many operating systems out there, but 
> I'm sure you can find ppl who are willing to compile it for you.
> Im no expert with *nix in general, and never even seen/used java as a 
> language, so my suggestion might not even be possible.
> 
> -Noam M.
> 
> 
> 
> -------------------------------------------------------
> This SF.NET email is sponsored by:
> SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
> http://www.vasoftware.com
> _______________________________________________
> Servicesng-chat mailing list
> Ser...@li...
> https://lists.sourceforge.net/lists/listinfo/servicesng-chat

Re: [SNG-chat] Let's start something up (was: Re: [epona] Epona "is this the end")

[Noam M. <no...@be...>]

Daniele Nicolucci (Jollino) wrote:
<snip>
> - Installing a recent JRE to make services live in a fast JVM would 
> probably be too tricky, if not impossible, in some configurations
> - Installing gcc 3.x to compile Java stuff into native machine code 
> would probably be even worse
> 
> This is because many machines running irc services are often used for 
> other stuff, and they just can't afford the risk of upgrading the JRE 
> (they could be using java for some stuff that can't be interrupted), or 
> even worse, upgrading to gcc 3.x to compile services. Upgrading gcc 
> could even mean, in the most Armageddon-like situation, that the whole 
> system ends up being messed up by accidental overwriting of libraries. 
> It's a possibility, although it's improbable. But it's still possible.
> 
I have a suggestion on how to solve this problem.
provide compiled versions of your services on all popular operating systems.
this might be tricky due to the so many operating systems out there, but 
I'm sure you can find ppl who are willing to compile it for you.
Im no expert with *nix in general, and never even seen/used java as a 
language, so my suggestion might not even be possible.

-Noam M.

[SNG-chat] Let's start something up (was: Re: [epona] Epona "is this the end")

[Daniele N. (J. <jo...@so...>]

(from the epona ml)

Gioved=EC, 23 gen 2003, alle 18:31 Europe/Rome, Lucas Nussbaum ha =
scritto:

> nobody with enough coding
> skills  willing to take the role of project leader. All the potential
> developpers are looking at each other waiting and thinking "I'll let
> them start, then I'll help". =3D> many super-contributors, but no =
coding
> team.

So, here we are again.
I think that the main reason for people not starting anything (apart=20
from a possible lack of time) is that no one knows exactly what=20
language this is going to use.

Lucas proposed Java, others didn't agree. His arguments about Java=20
being currently faster than the past, and machine-compilable, just=20
didn't convince others (I must admit I'm personally not sure).
So, I'm going to give my usual two cents about this thing, hoping that=20=

others will at least contribute in this discussion, which, we hope, is=20=

going to decide what language this stuff is going to be written in.=20
Otherwise I'll start doing something in Python myself. :)

---8<---
** Why Java would be good for services **
- Fast to develop
- Easy to mantain
- Compilable into machine code to run as fast as C stuff
- Runnable through a Java Virtual Machine

** Why Java would not be good for services **
- It could probably be slower than C
- Installing a recent JRE to make services live in a fast JVM would=20
probably be too tricky, if not impossible, in some configurations
- Installing gcc 3.x to compile Java stuff into native machine code=20
would probably be even worse

This is because many machines running irc services are often used for=20
other stuff, and they just can't afford the risk of upgrading the JRE=20
(they could be using java for some stuff that can't be interrupted), or=20=

even worse, upgrading to gcc 3.x to compile services. Upgrading gcc=20
could even mean, in the most Armageddon-like situation, that the whole=20=

system ends up being messed up by accidental overwriting of libraries.=20=

It's a possibility, although it's improbable. But it's still possible.

** Alternatives **
The most praised alternative to Java would be C++, even though C++=20
would be a PITA(TM). It inherits all the bad things of C, multiplied by=20=

the tricky use of classes. This means that coders would lose=20
significantly more time playing around with pointers and references=20
than writing real code. This isn't very productive, but could be eased=20=

by writing a good interface base to the internal services system. Maybe=20=

C would make servicesng a bit easier to develop than C++.

Other possible alternaives that I could think of would be interpreted=20
languages, such as Perl of Python, but I'm not taking them into the=20
discussion because most people would think that they wouldn't fit. I=20
mean, they'd probably work fine anyway, since we're not computing prime=20=

numbers or something like that, but the only way to see how good (or=20
bad) they scale is just a matter of checking AFTER the whole thing is=20
done.

Anyway, between Perl and Python I'd choose Python without a second=20
thought. It's easier to read, to develop and to mantain, and I could=20
help myself. And we could write the killer functions in C for the sake=20=

of speed.

Anyone interested, have a look at=20
<http://www.linuxjournal.com/article.php?sid=3D3882>.
--->8---

Regards
--=20
Daniele Nicolucci (Jollino) - jo...@so...
IRC Operator on Discussioni.Org (www.discussioni.org)

[SNG-chat] Skimming through Lucas' files

[Daniele N. (J. <jo...@so...>]

> There might be one or several config files (one per module ?). Those 
> should be
> organized in a hierarchical way.

I think that there should be a single config file to be read by the 
Config module, which could contain some include directives to split the 
config into smaller files. This would let anyone choose their way to 
configure Services, e.g. having a single big file or moving something 
out of it.

> The config files should be written in an easy to modify format (XML ?)

Maybe a simpler (i.e. easier to read and understand) syntax could be 
preferred. Something on the line of Epona's conf, maybe?

> The logger should be the first thing to be started. On start, it 
> should only
> use the StdOut output. Other outputs should be added after the config 
> files
> are read.

By default, the stdout output should be as concise as possible when 
Services are not running in a debug environment. This means that it 
could write an error message if something goes wrong, otherwise 
silently fork into background. Maybe this should be changed via a 
command line option?
And how about logging to stderr by default?

> 6. Language Abstraction Layers
> They handle the internationalization of services. How this will be 
> done is
> undetermined for now.

No idea yet, but I fear that the whole text stuff should be loaded in 
memory. Maybe Services could dynamically unload language packs that 
aren't used by any user, and possibly load them back when someone needs 
them.
For the sake of memory allocation, the config file(s) should allow to 
list the language packs that must always stay in memory.

> - It seems difficult to create a generic DB abstraction layer (that 
> every
>   SModule author could use transparently). It seems more likely that 
> every
>   SModule will have its own DB abstraction layer. But several SModules 
> could
>   still share the same. This is still to be determined.

At least with SQL DBMSs, the stuff is quite similar.
I'm no good at it so I'd better shut up 'fore saying crazy things.


My useless 2 cents about the whole stuff. :)
-- 
Daniele Nicolucci (Jollino) - jo...@so...
Web applications programmer (www.webdreamers.net)
IRC Operator on Discussioni.Org (www.discussioni.org)
Administrator of the Lingua Mundi project, www.linguamundi.sogno.net

Re: [SNG-chat] Global Organization diagram, List of subsystems, Commits mailing list

[Daniele N. (J. <jo...@so...>]

Venerd=EC, 10 gen 2003, alle 15:30 Europe/Rome, Lucas Nussbaum ha =
scritto:

> I posted those 3 documents on tuesday, asking for comments and
> questions, so that we can discuss them. There has been none.

I'm personally not good about this kind of things, and I'm not going to=20=

be a code developer for this project due to my very little knowledge=20
and self-confidence ;)
I've read that stuff, but I really didn't know what to reply about.

Everything's fine for me, provided that the thing will work without too=20=

much hassle (and we're going to back to the java/nojava discussion here=20=

:)

> Other developpers, please take your
> responsabilities, make initiatives, etc !

I've been adding some stuff now and then to the 'features' file in the=20=

cvs tree...
--=20
Daniele Nicolucci (Jollino) - jo...@so...
Web applications programmer (www.webdreamers.net)
IRC Operator on Discussioni.Org (www.discussioni.org)
Administrator of the Lingua Mundi project, www.linguamundi.sogno.net

Re: [SNG-chat] Global Organization diagram, List of subsystems, Commits mailing list

[Lucas N. <lu...@lu...>]

Hi,

I posted those 3 documents on tuesday, asking for comments and
questions, so that we can discuss them. There has been none.

I'd like to outline something :
- I won't code this project alone : if nobody is willing to help, I
  won't do it.
- I would prefer not to be the project leader. And I would prefer not to
  have to act as a project leader. Other developpers, please take your
responsabilities, make initiatives, etc !

Thanks

lucas

Re: [SNG-chat] Modules idea...

[<nha...@gm...>]

----- Original Message -----
From: Daniel Engel <da...@ze...>
Date: Tuesday, January 7, 2003 9:12 am
Subject: [SNG-chat] Modules idea...

>   that way all
>   modules (at least those with pseudo-clients) run on their own
>   process. I guess you could even get fancy and have backup 
>   modules (i.e. if NickServ dies, the backup takes over, and the
>   returning stays as a backup).

I'll throw in my change from a different point of view. 

A lot of people pay for a shell, and are allowed to run an IRCd, and 
one services package. For the most part, the company will check to make 
sure you are only running 2 proccesses on your shell. Sometimes they'll 
allow 3-4 proccesses. Either way though, for those people who aren't 
running the IRCd's on their own boxes, their service providers usually 
(from my experiences) limit the number of seperate proccesses they can 
run.

So, by running each module as a seperate processes, it's potentially 
limiting the number of people that can run these services...

Anyways.. That's just a thought...

Re: [SNG-chat] Modules idea...

[David R. <dav...@vi...>]

meow,

re: treating services as clients, that connect - imho the socket side
isn't an issue, well not compared to the problem as multiple processes
running on a shell - this is a big concern to a lot of server admins, as
many shell providers only allow a limited number of processes, if services
needed say 5 processes many people could simply not run it.

re: modules - as for modules being a replacement for a lack of classes i
don't agree, what i would like to see on a module basis is the inclusion
of a config file, which specifies a path to the module location, this
module can then be loaded and work, ideally do-able without a services
restart, and most defiantly without anyone needing to change the core
source code.

anyway, figured i would show im still alive :)

Rob

Re: [SNG-chat] Modules idea...

[Lucas N. <lu...@lu...>]

On Tue, Jan 07, 2003 at 09:12:15AM -0500, Daniel Engel <da...@ze...> wrote:
> I've been thinking about modules lately...
> 
>    One of the reasons I didn't like NeoStats module handling,
>    is that if a module segfauls, the entire service process
>    comes down. One solution would be to put every module on 
>    their own threads, but I thought... 

Mmmh I'm not sure this is a good idea.
It would mean that modules would communicate with the core through
sockets. we would have to define a protocol, and encode/decode
messages. this might cause performance problems (and performance seems
to be a problem for many people here).
Java, if chosen, would provide a nice solution : many errors are catched
by exceptions.

I'm not sure we need real "modules" anyway. Modules is a way to
circumevent the lack of object oriented things in classic languages like
C. I think the only point is : Can a third party code additionnal
classes for services without changing anything in the services code ? If
we can reach this point, this would be enough modularity for me.

For the backup problem, I don't think a backup system is needed for such
software crashes. The only times when I have needed backup services were
due to connection or hardware problems where the services were hosted.
And, with such problems, backup services should be started by hand
anyway. The only problem is to make databases be up to date on the
backup machine. This can be done by replication of the SQL databases, or
rsync. 

lucas

Re: [SNG-chat] Modules idea...

[Daniele N. (J. <jo...@so...>]

Marted=EC, 7 gen 2003, alle 15:12 Europe/Rome, Daniel Engel ha scritto:

>    What if the modules connect to the service server as a client?
>    Is that feasable? We can take care of ping timeouts, and keep
>    a fairly steady connection trough the localhost.

This means that the main services process (the core) would need to=20
listen on a port to accept incoming connections... I'm not sure if this=20=

would be better than having services running on different child=20
processes of the parent - core - process.

Anyway, such a client/server architecture could be implemented in a=20
more personal way, without necessarily resembling the way irc clients=20
connect to irc servers. Of course, the core will need to have the=20
standard irc support cabality to communicate with its real uplink, but=20=

modules would not need it.

My usual 2 crazy cents
--=20
Daniele Nicolucci (Jollino) - jo...@so...
Web applications programmer (www.webdreamers.net)
IRC Operator on Discussioni.Org (www.discussioni.org)
Administrator of the Lingua Mundi project, www.linguamundi.sogno.net

[SNG-chat] Modules idea...

[Daniel E. <da...@ze...>]

I've been thinking about modules lately...

   One of the reasons I didn't like NeoStats module handling,
   is that if a module segfauls, the entire service process
   comes down. One solution would be to put every module on 
   their own threads, but I thought... 

   What if the modules connect to the service server as a client?
   Is that feasable? We can take care of ping timeouts, and keep
   a fairly steady connection trough the localhost. that way all
   modules (at least those with pseudo-clients) run on their own
   process. I guess you could even get fancy and have backup 
   modules (i.e. if NickServ dies, the backup takes over, and the
   returning stays as a backup).

   I guess my question is if it is good idea. Can some of you 
   give a though and share your findings? This model has not
   been inplemented on any services to date, so we can't do
   any actual comparisons...

d.
   

[SNG-chat] Global Organization diagram, List of subsystems, Commits mailing list

[Lucas N. <lu...@lu...>]

Hi,

Global Organization diagram :
-----------------------------
Using the features list, I tried to build an architecture which will be
able to deal with everything we planned. I committed the diagram[1] to the
CVS repository, and explained everything in a text file[2]. If you
modify the diagram, use the .dia file in the repos, and regenerate a png
file with dia.

[1] http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/*checkout*/servicesng/sng-design/globalorg.png?rev=HEAD&content-type=image/png
[2] http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/*checkout*/servicesng/sng-design/globalorg.txt?rev=HEAD&content-type=text/plain

Please discuss this here. We need to have something very clean before
going any further.

Subsystems list :
-----------------
I also tried to determine how work on services could be split, with a
priority list. See http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/*checkout*/servicesng/sng-design/subsystems.txt?rev=HEAD&content-type=text/plain

Same here, please comment it.

Mailing list for CVS commits :
------------------------------
I created a mailing list to log CVS commits. You can subscribe on
http://lists.sourceforge.net/lists/listinfo/servicesng-commits . Nobody
(except me) has been subscribed yet.
-- 
Lucas Nussbaum

Re: [SNG-chat] hello there?

[Lucas N. <lu...@lu...>]

On Tue, Jan 07, 2003 at 02:16:21PM +0100, Daniele Nicolucci <jo...@so...> wrote:
> Martedì, 7 gen 2003, alle 08:12 Europe/Rome, Lucas Nussbaum ha scritto:
> 
> >Once again, please don't compare
> >Java 1.1 in IE with Java 2 as implemented in JDK 1.4. I'm sure Java 
> >will
> >be fine for our services.
> 
> Lucas, the problem is: we all know that services are possibly going to 
> run on already stable machines that could maybe handle a lot of other 
> kinds of services for paying clients; this means that these machines 
> cannot take the risk to mess something up to install a Java 2 VM and/or 
> gcc 3.0x to compile the java sources into standard executables.
> And personally, nor would I personally like to run precompiled stuff.

Ok, this discussion will be held later. Let's do the design part, see
who will be coding, and let's vote to determine the language.
-- 
Lucas Nussbaum

Re: [SNG-chat] hello there?

[Daniele N. (J. <jo...@so...>]

Marted=EC, 7 gen 2003, alle 08:12 Europe/Rome, Lucas Nussbaum ha =
scritto:

> Once again, please don't compare
> Java 1.1 in IE with Java 2 as implemented in JDK 1.4. I'm sure Java=20
> will
> be fine for our services.

Lucas, the problem is: we all know that services are possibly going to=20=

run on already stable machines that could maybe handle a lot of other=20
kinds of services for paying clients; this means that these machines=20
cannot take the risk to mess something up to install a Java 2 VM and/or=20=

gcc 3.0x to compile the java sources into standard executables.
And personally, nor would I personally like to run precompiled stuff.

Just my two cents...
--=20
Daniele Nicolucci (Jollino) - jo...@so...
Web applications programmer (www.webdreamers.net)
IRC Operator on Discussioni.Org (www.discussioni.org)
Administrator of the Lingua Mundi project, www.linguamundi.sogno.net

Re: [SNG-chat] hello there?

[Mathieu A. <ma...@ma...>]

--On lundi 6 janvier 2003 14:16 -0500 Daniel Engel <da...@ze...> wrote:

> I don't think we should starting two code bases, if
> you'd like to go with Java... let's do java. 

I don't believe that java is the best language for services, but... :)

-- 
Mathieu Arnold

Re: [SNG-chat] hello there?

[Lucas N. <lu...@lu...>]

On Mon, Jan 06, 2003 at 02:16:56PM -0500, Daniel Engel <da...@ze...> wrote:
> I don't think we should starting two code bases, if
> you'd like to go with Java... let's do java. 

Ok. Anyway, we can still do all the design part on a language
independent state. If, after we step, you think it's doable in C++
without taking the risk not to ever release something because of
complexity, we'll discuss that again.

Most, if not all of the griefs expressed here about Java are typical
results from FUD of compagnies like MS. Once again, please don't compare
Java 1.1 in IE with Java 2 as implemented in JDK 1.4. I'm sure Java will
be fine for our services.

I've started to design SNG's global architecture yesterday. I'll upload
the result later today : some things still have to be fixed.

Lucas

Re: [SNG-chat] hello there?

[Daniel E. <da...@ze...>]

I don't think we should starting two code bases, if
you'd like to go with Java... let's do java. 

d.

On Mon, Jan 06, 2003 at 05:22:01PM +0100, Lucas Nussbaum wrote:
> On Mon, Jan 06, 2003 at 12:39:50PM +0100, Daniele Nicolucci <jo...@so...> wrote:
> > hey there,
> > are you all dead? :)
> > 
> > do not forget about this project ... :)
> 
> Hi,
> 
> I got back from holidays yesterday.
> First of all, sorry : I didn't have time to do a UML class diagram of
> SNG. But I thought a lot about it.
> 
> I don't think that I know C++ well enough to use it to code services.
> If Daniel and others want to start something with C++, I think the best
> thing to do is :
> 1) We decide of a common architecture for both C++ and Java
> 2) I (and others) start something in Java, while Daniel (and others)
> start something in C++.
> 
> Who amonst you is interested in coding SNG in Java ? (Please reply to
> this mail, either on the list or privately, and describe your knowledge
> of java)
> 
> Who, amonst those who want to code but don't know Java, is interested in
> learning Java during this project ? (Please reply and describe your
> object-oriented programming languages experience, if any)
> 
> Thanks,
> 
> lucas
> 
> 
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Servicesng-chat mailing list
> Ser...@li...
> https://lists.sourceforge.net/lists/listinfo/servicesng-chat

Re: [SNG-chat] hello there?

[Lucas N. <lu...@lu...>]

On Mon, Jan 06, 2003 at 12:39:50PM +0100, Daniele Nicolucci <jo...@so...> wrote:
> hey there,
> are you all dead? :)
> 
> do not forget about this project ... :)

Hi,

I got back from holidays yesterday.
First of all, sorry : I didn't have time to do a UML class diagram of
SNG. But I thought a lot about it.

I don't think that I know C++ well enough to use it to code services.
If Daniel and others want to start something with C++, I think the best
thing to do is :
1) We decide of a common architecture for both C++ and Java
2) I (and others) start something in Java, while Daniel (and others)
start something in C++.

Who amonst you is interested in coding SNG in Java ? (Please reply to
this mail, either on the list or privately, and describe your knowledge
of java)

Who, amonst those who want to code but don't know Java, is interested in
learning Java during this project ? (Please reply and describe your
object-oriented programming languages experience, if any)

Thanks,

lucas

[SNG-chat] hello there?

[Daniele N. (J. <jo...@so...>]

hey there,
are you all dead? :)

do not forget about this project ... :)
-- 
Daniele Nicolucci (Jollino) - jo...@so...
Web applications programmer (www.webdreamers.net)
IRC Operator on Discussioni.Org (www.discussioni.org)
Administrator of the Lingua Mundi project, www.linguamundi.sogno.net

[SNG-chat] Volunteer

[David M. <da...@ic...>]

Hey there,

Not really sure what to say, but I'd like to volunteer my services for this
project (even though it's a bit late). I saw the discussions going on in the
ircservices mailing list and agree that a ground-up approach is probably the
best idea, since most of the services out there at the moment are based from
the original ircservices code written quite a while ago.

I'm based in the UK and I'm currently operating an Unreal network with
ircservices 5 with about 120 users on at any one time. I've got quite a lot
of experience with coding in C/C++, but I'm not the best at it. If you want
to see an example of my code, you can check out
http://davep120.tripod.com/life3.cc.txt, which is a small piece of C++ code
for the game of life thing that I made a few months ago.

My suggestion would be for the services to use C++ instead of a C/Java
codebase. Most people have a C++ compiler now since (more or less) everyone
seems to use gcc, and C++ is a much more powerful language in my opinion.
Then again, I'm biased against Java because I've not really looked into it,
and I prefer C++ anyway :P

So, that's me. I'd be grateful if you did let me help out on the project a
lot, because I need the experience more than anything else.

Cheers,

------------------
nodnuke23 <da...@ic...>
icerealm.net network administrator
PGP key: http://icerealm.net/nodkey.pem
Fingerprint: FB66 A202 1637 4D73 81A5  F001 E660 499D 6494 C0FB

Re: [SNG-chat] Preferred UML tool

[<med...@un...>]

Am 20 Dec 2002 um 22:38 hat Lucas Nussbaum geschrieben:

It is free software, and is available on windows (see
> http://dia-installer.sourceforge.net/). It will probably be good
> enough for our needs. (there's even a dia2code that generates code
> from a dia UML class diagram :)
> 
> dia homepage : http://www.lysator.liu.se/~alla/dia/

thank you very much. i've been searching for such a tool since hours. i found only a lot 
of commercial tools for windows and mostly they depended in any way on java related 
things which don't run on my box :)

best regards

-- 
medusaXX
med...@un...

Re[2]: [SNG-chat] My choice : Java

[Rory <ro...@ca...>]

Hello Dan,

Saturday, December 21, 2002, 4:50:42 AM, you wrote:

DM> I might as well jump in here, seems like as a good a time as any.
DM> My name is Dan Morey and I'm 27 and have been using IRC since Feb. of 96 and I have been involved in
DM> several networks over the years. What I'd like to comment on is your list of the three kinds of
DM> people who run services.

>> People running services can be classified into 3 categories :
>> - people running services on a shell account. Those pay much for this,
>>   and can ask the shell provider to install packages for gcc 3.2 or
>>   sun's jdk.
>> - people running services on a machine they have root. Those can install
>>   gcc 3.2 or sun's jdk themselves.
>> - people running services on a machine they know the root personally, so
>>   they can ask.

DM> This is a fairly complete list but I strongly disagree with the first assumption. Shell companies do
DM> not really care about the client; people who want to run IRCd and services are a dime a dozen and
DM> usually more headache than they are worth because IRC networks attract DoS attacks. You would think
DM> that they would know this and if they cared that much not host IRCd plans. I recently ran into a
DM> problem on the shell we ran services on, we needed sendmail or qmail installed to use sendpass on
DM> Epona services, when we asked the admin to enable/install this we were told that there are 12 other
DM> people on that machine running IRCd and services and they hadn't asked for it so he wasn't going to
DM> do it. We ended up moving services to another machine and the admin is thinking about canceling her
DM> account.

DM> Granted a lot of networks do have at least one server that is linked where they have root or know
DM> someone who has root, but I really think that compatibility of existing systems/hardware should be
DM> seriously considered. I don't think you should assume that shell renters are going to be able to get
DM> the things that services need installed on the machines they use if they aren't there already. If it
DM> turns out to be a major headache people are just going to say screw it and find services that are
DM> easier to install.

DM> Just my two cents.

DM> Dan Morey

I would have to second that, Dan. I run Services on a shell account
(came as a package deal with the ircd account) and I know for a fact
that very few, if any, providers would give a damn about installing
something extra at the request of a customer. You takes what you've
signed up for or go elsewhere. No skin off their nose.



-- 
Best regards,
Rory                            
mailto:ro...@ca...

[SNG-chat] Preferred UML tool

[Lucas N. <lu...@lu...>]

Hi,

During my holidays, I'll start to design an UML class diagram for the
services. I'll do so using dia.
dia is a GTK application used to draw diagrams. It is free software, and
is available on windows (see http://dia-installer.sourceforge.net/). It
will probably be good enough for our needs. (there's even a dia2code
that generates code from a dia UML class diagram :)

dia homepage : http://www.lysator.liu.se/~alla/dia/

Lucas

Re: [SNG-chat] My choice : Java

[Daniele N. (J. <jo...@so...>]

Venerd=EC, 20 dic 2002, alle 20:25 Europe/Rome, Daniel Engel ha scritto:

> Nice... X-Mailer: Apple Mail (2.551)
>
> I'm *this* close from buying the new 17" iMac...

Good choice, unless you're going to do graphics.
If you can, ask to try the iMac before actually buying it, since some=20
LCD iMacs have been shipped with broken pixels and Apple doesn't change=20=

the computer if there are 3 or less faulty pixels.
Other LCD screen vendors won't change them if there are 12 or less, so=20=

Apple is better in this sense - but why risking?

Right now I'm using my iBook ;)

I'd like to get an eMac with superdrive, since till the end of december=20=

there is an offer for students, but I can't afford it anyway right now.=20=

Sigh. Too bad...

Ok, offtopic enough. :)
--=20
Daniele Nicolucci (Jollino) - jo...@so...
Web applications programmer (www.webdreamers.net)
IRC Operator on Discussioni.Org (www.discussioni.org)
Administrator of the Lingua Mundi project, www.linguamundi.sogno.net