1

I tried to get the read-only svn using several different computers.... one with debian, one running freebsd, and a windows computer from all of them i get errors like this:

svn: Failed to load module for FS type 'fsfs'


I tried registering for the trac tool, but the recaptcha is not working. I tried with both firefox and IE. I get this message: "You did not enter the ReCaptcha verification corecly." Also, since I'm complaining smile corecly should be correctly wink

I have an interest in adding Flash App support, but I'm not sure if it's going to move beyond the idea stage at this point as I have lots of other things going on right now... smile

Greg

2

Why not work with TIGCC CVS HEAD, which actually checks out properly?

FlashApp support needs work both in ld-tigcc and in TIGCCLIB. ld-tigcc needs to handle the format and either create the headers or auto-import them from tigcc.a (like the kernel and FlashOS headers work). TIGCCLIB needs changes to the startup code (the whole concept of "startup code" probably isn't really appropriate for FlashApps, plus a lot of code there is self-modifying or relocates the executable, neither of which will work in FlashROM, and finally it may make sense to put some headers into tigcc.a), a way to fill in the FlashApp header (AFAIK, some fields are supposed to be settable by the application) and changes to those functions which are self-modifying in one way or the other (some TIGCCLIB functions construct a jmp inside the code, some have global variables inside the code section).

Just write the necessary patches against TIGCC CVS HEAD and send them to me, that way they'll end up in the official TIGCC, and I'm sure GCC4TI will be importing them from there too, they're quite effective at stealing my work.
avatar
Mes news pour calculatrices TI: Ti-Gen
Mes projets PC pour calculatrices TI: TIGCC, CalcForge (CalcForgeLP, Emu-TIGCC)
Mes chans IRC: #tigcc et #inspired sur irc.freequest.net (UTF-8)

Liberté, Égalité, Fraternité

3

!call Godzil
--- Call : Godzil appelé(e) sur ce topic ...

Now that you've fixed the Trac, please fix the SVN too wink
(perhaps a simple restart of the daemon will do, as it did for Apache grin)
I'm sure GCC4TI will be importing them from there too, they're quite effective at stealing my work.

What are you exactly trying to do with this kind of statements, Kevin ? Pissing us off so we ban you from this board ? No way, sorry wink
avatar
Membre de la TI-Chess Team.
Co-mainteneur de GCC4TI (documentation en ligne de GCC4TI), TIEmu et TILP.
Co-admin de TI-Planet.

4

Sorry I always forget to restart the svnserver after each update... That's corrected now, sorry for the convienience...
avatar
Proud to be CAKE©®™


GCC4TI importe qui a problème en Autriche, pour l'UE plus et une encore de correspours nucléaire, ce n'est pas ytre d'instérier. L'état très même contraire, toujours reconstruire un pouvoir une choyer d'aucrée de compris le plus mite de genre, ce n'est pas moins)
Stalin est l'élection de la langie.

5

cool, but i still can't register with trac ... it doesn't think i'm getting the capatcha right.

I have no interest in being involved in politics. I'll try to work with both projects if I get to the point where I have something to contribute. I actually didn't know that tigcc has cvs...

Greg

6

TIGCC has had a CVS repository ever since Sebastian left and I (as the only remaining developer) decided to do the development more openly (and also reduce the risk of catastrophic data loss). All the post-0.95 work was done in CVS.

Unlike what the GCC4TI folks repeatedly claim, I have always been and still am for open development (more so than GCC4TI which kept their SVN repository private until their first release, and whose developers still from time to time develop in a private git branch and only commit to SVN in huge code dumps hours before the release).
avatar
Mes news pour calculatrices TI: Ti-Gen
Mes projets PC pour calculatrices TI: TIGCC, CalcForge (CalcForgeLP, Emu-TIGCC)
Mes chans IRC: #tigcc et #inspired sur irc.freequest.net (UTF-8)

Liberté, Égalité, Fraternité

7

You don't seem to have noticed that Greg stated in ./4 that he has no interest in being involved in politics, do you ? I thought it was pretty clear. And you already knew about it.


You didn't reply to my question, so I'm going to ask again:
What are you exactly trying to do with this kind of statements, Kevin ? Pissing us off so we moderate your posts or even ban you from this board ? No way, sorry wink

You use very strict moderation rules on TIGCC board. Under such rules, the kind of statements you're posting would already had earned you a ban from this section months ago (well, actually, you'd already have been banned from yAronet since 2000).


For the record, since you seem to be determined at polluting the GCC4TI board: as for a proof to the openness of TIGCC, suffice it to mention the fact that TIGCC's bug list and todo list have been kept private for years. This led me to recently independently rediscover a bug you already knew about...
Oh, and BTW: if we're making "huge code dumps" (of peer-reviewed-and-improved code, something that you can't do because you're alone), that means that we're actually working on the code (build system improvements, etc.). Unlike you.
avatar
Membre de la TI-Chess Team.
Co-mainteneur de GCC4TI (documentation en ligne de GCC4TI), TIEmu et TILP.
Co-admin de TI-Planet.

8

It would be smart for both groups to have one git repository and maintain their own branches - that would help me out.... Then I could have my own "branch" and merge with both code bases easily. Sometimes people have to agree to disagree wink As long as it seems that both projects will be occasionally taking patches from each other, might as well make it easy.

still can't register on http://trac.godzil.net/gcc4ti/register

Greg

9

!call Godzil
--- Call : Godzil appelé(e) sur ce topic ...
I've just made multiple trials of account registration, and all of them failed, although I'm sure that I got several of them OK. JavaScript was enabled.


I've used Git(-svn) for all serious work (as opposed to a one-shot or few-shoots checkout(s) in a VM for testing, I mean) I performed on multiple work-time and free-time projects in the past two years, starting when I contributed some janitorial patches to Wine.
I'm definitely not against creating a Git repo for GCC4TI, all the more I had created a Git mirror for the TIGCC CVS repository.

In fact, the reason why GCC4TI hasn't been using Git from the beginning, is that in August 2008, more votes went to SVN than to Git, because:
* SVN is "good enough" in the sense that it fixes the most glaring design and implementation flaws of the old CVS crap;
* SVN has better interoperability with other SCMs, starting with Git;
* Git had, and probably still has despite improvements in all those areas, lesser tooling, lower portability and steeper learning curve than SVN.
Just like the initial decision on what SCM should be used (I'd have used Git if it had been just me grin), I feel that a complete switch from SVN to Git is not a decision I can make alone.
avatar
Membre de la TI-Chess Team.
Co-mainteneur de GCC4TI (documentation en ligne de GCC4TI), TIEmu et TILP.
Co-admin de TI-Planet.

10

There's no way I'm going to share a repository with GCC4TI. As far as I'm concerned, that project "does not exist", i.e. TIGCC will continue as if it didn't. GCC4TI is a rogue fork which does not respect the basics of meritocracy.
avatar
Mes news pour calculatrices TI: Ti-Gen
Mes projets PC pour calculatrices TI: TIGCC, CalcForge (CalcForgeLP, Emu-TIGCC)
Mes chans IRC: #tigcc et #inspired sur irc.freequest.net (UTF-8)

Liberté, Égalité, Fraternité

11

PpHd, Godzil, Folco, Flanker and me, between others, know how collaboration and meritocracy works: it's a fact, acknowledged by the Credits of dozens of programs, and hundreds of topics on message boards, that we've successfully applied it over the years with many persons.
Remember, collaboration and meritocracy is how I became a member of TICT: after multiple idea and code contributions of mine to TICT projects, Tom asked me whether I wanted to became a member.

If we didn't apply it in TIGCC, it's not our fault, sorry wink
We've already written many times that we got soooo tired up by most of our ideas and wished being rejected as a whole by you, the sole remaining member of TIGCC, that we gave up. When you announced completely irrealistic todo list items (on which you have made zero progress, more than one year and a month later, and counting !) while there are much more pressing issues on TIGCC (on a side note, hardly any of those have been tackled by you - of your six commits in TIGCC since January 2009, four are GCC4TI backports that tackle some of them, and the two others are trivialities), we said "let's do something together".
That's what we did, and we left you alone in the dust.
If it wasn't enough, several months later, you've tried to turn TIEmu and TILP into a dictatorship (by flat out refusing to cooperate with someone who went in through meritocracy), thereby negating all of Julien's and Romain's good work and good relationship towards the community - and added yourself a lot of workload and unneeded work by forking them. Good job.

All of this we've already written on other occasions, BTW, but you keep tirelessly spitting the same old crap...

Oh, BTW: you still haven't replied to a question I've already asked you twice before this one:
What are you exactly trying to do with this kind of statements, Kevin ? Pissing us off so we moderate your posts or even ban you from this board ? No way, sorry wink

I thought it was incorrect to ignore questions, requests posted by moderators (well, at least, I think so, all the more you hate it on the dead TIGCC/TICT board, when someone doesn't reply to you) ?
avatar
Membre de la TI-Chess Team.
Co-mainteneur de GCC4TI (documentation en ligne de GCC4TI), TIEmu et TILP.
Co-admin de TI-Planet.

12

Lionel Debroux (./11) :
We've already written many times that we got soooo tired up by most of our ideas and wished being rejected as a whole by you, the sole remaining member of TIGCC, that we gave up.

Why didn't you (all of you) ask yourself why your ideas and wishes got rejected? Or actually bothered reading my replies where I've always given the reason? You're just too proud to realize that not all your ideas are actually good ideas.
When you announced completely irrealistic todo list items (on which you have made zero
progress, more than one year and a month later, and counting !)

Such as? Getting KTIGCC working with kdewin isn't unrealistic at all and most likely we'd already be there if you guys helped with it instead of forking TIGCC and working on pointless things such as your unnecessary changes to the build scripts (e.g. changing #!/bin/bash to #!/bin/sh and then having to change stuff like echo -n; there was no bug there, as echo is a bash builtin, so echo -n is always available in a #!/bin/bash script!)…
while there are much more pressing issues on TIGCC

Such as? TIGCC 0.96 Beta 8 is very reliable, there are just a few minor bugs to fix.
(on a side note, hardly any of those have been tackled by you - of your six commits in TIGCC since January 2009, four are GCC4TI backports that tackle some of them, and the two others are trivialities), we said "let's do something together".

Why not together with me? Because I didn't let one of you become the head of the project out of the blue? Sorry, things don't work like that, as a new contributor, you start by sending patches, then you get commit after approval and only when you earned the trust of the existing contributors, you can get full comaintainership. (But of course, after what you did, none of you will ever get that level of trust. You should have thought of the consequences of your actions before doing them.)
That's what we did, and we left you alone in the dust.

You are the ones left alone in the dust, TIGCC is the real TIGCC, GCC4TI is just crap nobody uses.
If it wasn't enough, several months later, you've tried to turn TIEmu and TILP into a dictatorship (by flat out refusing to cooperate with someone who went in through meritocracy), thereby negating all of Julien's and Romain's good work and good relationship towards the community - and added yourself a lot of workload and unneeded work by forking them. Good job.

That's a complete misrepresentation of what happened to those projects. What did happen is that you got Julien and Romain to add you after sending a few minor contributions. (IIRC, the patches you sent before getting commit access can be counted on one hand, and they were all fairly small changes, some even buggy and incomplete (the IMG_INFO changes, which you've still not fixed, by the way).) Those folks never bothered to ask the other existing contributors (Tyler and me) for whether they trust you (hint: we don't, see above).

And then Julien and Romain not accepting me as Romain's successor even though I've been the #2 contributor after him for years, and also ignoring Tyler (the original author of the Group File Manager) was just too much. We still see ourselves as the legitimate successors of the old LPG, your projects as forks and you as usurpators of the project names.
What are you exactly trying to do with this kind of statements, Kevin ? Pissing us off so we moderate your posts or even ban you from this board ? No way, sorry wink

I'm just telling the truth.
avatar
Mes news pour calculatrices TI: Ti-Gen
Mes projets PC pour calculatrices TI: TIGCC, CalcForge (CalcForgeLP, Emu-TIGCC)
Mes chans IRC: #tigcc et #inspired sur irc.freequest.net (UTF-8)

Liberté, Égalité, Fraternité

13

We've already written many times that we got soooo tired up by most of our ideas and wished being rejected as a whole by you, the sole remaining member of TIGCC, that we gave up.

Why didn't you (all of you) ask yourself why your ideas and wishes got rejected? Or actually bothered reading my replies where I've always given the reason?

We all read, for years, why you rejected our ideas and wishes, sometimes in an ungraceful way (like, refusing to upgrade kernel support invoking the ideological "kernels suck !" pretext). And we still disagree with you, with technical or usage arguments to back up our ideas.
Again, I'll point you to my quote of MaeyHem's post on the TIGCC/TICT message board: topics/115787-gcc4ti-manifesto#5 . You have arguments to back up your claims, we have arguments to back up ours.
You're just too proud to realize that not all your ideas are actually good ideas.

You're just too proud to accept that there's an acceptable way besides your own ideas. As demonstrated by your numerous posts along the lines of "it sucks because it's proprietary" (sometimes posted in a completely off-topic way, you got kicked from several topics...) in various topics of the sections/2-3-coup-de-gueule , sections/2-4-jai-rien-a-dire-mais-jai-envie-den-parler , sections/2-1-discussions-generales sections anyway.
When you announced completely irrealistic todo list items (on which you have made zero
progress, more than one year and a month later, and counting !)

Such as? Getting KTIGCC working with kdewin isn't unrealistic at all

No matter the advent of KDE 4.3 (which is what 4.0, or at best 4.1, should have been if KDE had used a sane versioning scheme), forcing people to install the huge non-native kdewin bloat just in order to use a tiny IDE with few distinctive features, questionable capabilities and well-known deficiencies, remains a questionable idea. At the very least, it should definitely not be a priority task (something you gave the impression of, at the time).
and most likely we'd already be there if you guys helped with it instead of forking TIGCC and working on pointless things such as your unnecessary changes to the build scripts (e.g. changing #!/bin/bash to #!/bin/sh and then having to change stuff like echo -n; there was no bug there, as echo is a bash builtin, so echo -n is always available in a #!/bin/bash script!)…

We've already discussed that on #tigcc: reliance on bash is a portability bug, just as assuming that "make" is GNU make and "install" is GNU install. In the install scripts, we've fixed those portability bugs (which was trivial) and more, added proper error handling, improved the packaging, and more.

while there are much more pressing issues on TIGCC
Such as? TIGCC 0.96 Beta 8 is very reliable, there are just a few minor bugs to fix.

TIGCC 0.96 Beta 8 is nearly three years old (!!), there have been many fixes in the TIGCC CVS and GCC4TI SVN repository since then.
Though we tackled several issues and closed the corresponding tickets, the GCC4TI Trac ticket list is a testimony to the range and depth of open issues on TIGCC... No wonder, because hardly anything new happened in TIGCC since 2005 or so, despite the existence of most of that huge todo/wish list.
(on a side note, hardly any of those have been tackled by you - of your six commits in TIGCC since January 2009, four are GCC4TI backports that tackle some of them, and the two others are trivialities), we said "let's do something together".
Why not together with me?

See above and in many other topics. We've been able to cooperate with dozens of persons in the community over the years, but we haven't been able to cooperate with you, due to your lack of respect towards our ideas and our persons...
You like people as long as they agree with you 100%, but your relationships with other people deteriorate if the agreement rate falls.
Because I didn't let one of you become the head of the project out of the blue? Sorry, things don't work like that, as a new contributor, you start by sending patches, then you get commit after approval and only when you earned the trust of the existing contributors, you can get full comaintainership. (But of course, after what you did, none of you will ever
get that level of trust. You should have thought of the consequences of your actions before doing them.)

By rejecting many of our ideas, you didn't even let us a chance to make and send most patches (because they wouldn't have gotten in anyway).
That's what we did, and we left you alone in the dust.
You are the ones left alone in the dust, TIGCC is the real TIGCC, GCC4TI is just crap nobody uses.

That's both wishful thinking (contradicted only by facts) and a show of total disrespect towards people's work and ideas. But oh well, we on yAronet are accustomed to it. You've been like that for years, and it took way too long to me to understand that.
What did
happen is that you got Julien and Romain to add you after sending a few minor contributions.

THAT is a misrepresentation of the facts: as already written elsewhere, and acknowledged by Romain, I never asked for commit rights.
(IIRC, the patches you sent before getting commit access can be counted on one hand, and they were all fairly small changes, some even buggy and incomplete (the IMG_INFO changes, which you've still not fixed, by the way).)

In case you hadn't noticed, in the past few weeks, I've spent a lot of time on RSA key factoring and the rest. A community adventure of which you've been totally missing.
As one of the co-maintainers at the time, you were Cc on the patches, so you're also to blame if an incomplete change went in. As one of the co-maintainers for a long time, you're also to blame for the code duplication (that's the root cause of the problem - no code duplication, no way to miss the modification of one of the copies) I didn't detect.
Those folks never bothered to ask the other existing contributors (Tyler and me) for whether they trust you (hint: we don't, see above).

Do you really think they had any need for asking the most biased and predictable person they know of ?
And then Julien and Romain not accepting me as Romain's successor even though I've been the #2 contributor after him for years, and also ignoring Tyler (the original author of the Group File Manager) was just too much.

As you briliantly demonstrated in topics/122111-tilptiemu-arret-du-developpement (auto-proclamation of maintainership + "I am the maintainer, I do what I want"), you were going to lock down the traditionally open-minded LPG software the way you locked down TIGCC. This was totally unacceptable for the sake of the community - and an utter disrespect for Julien and Romain, who have spent lots of time and energy into raising those projects to what they are. They didn't want to see their effort misappropriated by a well-known sabotager.
We still see ourselves as the legitimate successors of the old LPG, your projects as forks and you as usurpators of the project names.

See yourselves the way you wish wink
What are you exactly trying to do with this kind of statements, Kevin ? Pissing us off so we moderate your posts or even ban you from this board ? No way, sorry wink
I'm just telling the truth.

LOL, I was sure that you were going to blurt out that obscenity rotfl
You're just telling what you'd like the truth to be, and what you'd like people to believe. That's very different.
avatar
Membre de la TI-Chess Team.
Co-mainteneur de GCC4TI (documentation en ligne de GCC4TI), TIEmu et TILP.
Co-admin de TI-Planet.

14

Lionel Debroux (./13) :
refusing to upgrade kernel support invoking the ideological "kernels suck !" pretext

The actual reason (and the one I've always given) is that using PreOs-only features makes programs crash when run on older kernels, and there's no reliable way to check for PreOs or a derivative (there are at least 4 different identifiers: PreOs, PedroM, TitaniK, Iceberg, and there may be more in the future, and there are also version numbers to check for most stuff, which are different for each of those).
No matter the advent of KDE 4.3 (which is what 4.0, or at best 4.1, should have been if KDE had used a sane versioning scheme), forcing people to install the huge non-native kdewin bloat just in order to use a tiny IDE with few distinctive features, questionable capabilities and well-known deficiencies, remains a questionable idea. At the very least, it should definitely not be a priority task (something you gave the impression of, at the time).

It's a priority task because it allows adding features to the IDE without having to code them twice (of which once in a proprietary language I don't even have a compiler for). Code duplication is bad, as you say yourself.
Though we tackled several issues and closed the corresponding tickets, the GCC4TI Trac ticket list is a testimony to the range and depth of open issues on TIGCC...

Most of them are non-issues like the use of bash in the install scripts (which is really a feature). Some are actual bugs, but rarely encountered in practice.
See above and in many other topics. We've been able to cooperate with dozens of persons in the community over the years, but we haven't been able to cooperate with you, due to your lack of respect towards our ideas and our persons...You like people as long as they agree with you 100%, but your relationships with other people deteriorate if the agreement rate falls.

Quite funnily, that's about what I think of you… You only honor existing project structures when it suits you, you don't hesitate to putsch your way into projects you don't like the leader of (see what you did to TIGCC (where at least you had to change the name) and LPG (where you're usurpating the original project's name, helped by Julien and Romain who decided to support you rather than their existing codevelopers (Tyler and me) based on nationality)).
By rejecting many of our ideas, you didn't even let us a chance
to make and send most patches (because they wouldn't have gotten in anyway).

Patches for actual bugs would definitely have been accepted, unless the "fix" is broken and causes worse issues. On one hand, you claim to tackle actual issues in TIGCC, on the other hand, you say you forked it because you don't like my decisions, now which is it?
THAT
is a misrepresentation of the facts: as already written elsewhere, and acknowledged by Romain, I never asked for commit rights.

But you surely didn't complain. wink You also didn't offer to step down when I told you I don't trust you and I refuse to work with you. But I blame Julien and Romain most of all, they should have asked me and Tyler before granting you access.
In case you hadn't noticed, in the past few weeks, I've spent a lot of time on RSA key factoring and the rest. A community adventure of which you've been totally missing.

A completely useless task as there have been ways to bypass the whole signature process, and also a legal risk.
As one of the co-maintainers at the time, you were Cc on the patches, so you're also to blame if an incomplete change went in.

Well, Romain expeditedly committed your patch and rushed the release at a time where I was extremely busy with other stuff (university and/or Fedora stuff, probably both) and had little time to review it. I thought he had and made the mistake to trust him on that, he obviously didn't. roll (Code quality has never been his strength, I used to always proofread his commits and I caught several mistakes that way.)
As one of the co-maintainers for a long time, you're also to blame for the code duplication (that's the root cause of the problem - no code duplication, no way to miss the modification of one of the copies) I didn't detect.

That "duplicate code" has been written by Romain before I even joined the project, he's the one to blame there. That said, it's not really duplicate code, it's just 2 pieces of code who happen to use the same structure. It's you who changed the structure without even grepping for it in the whole source code.
Do you really think they had any need for asking the most biased and predictable person they know of ?

I was the #2 developer, it was extremely impolite to tell me only after the fact that they accepted a new developer.
auto-proclamation of maintainership

For me, it was completely logical that I was and I didn't even conceive anything else. The succession rules in Free Software projects are clear, if the #1 developer leaves, the #2 developer becomes the #1 developer. So it wasn't so much auto-proclamation as just a statement of what I thought to be undisputed facts.
"I am the maintainer, I do what I want"

That's also how things work in the real world.
you were going to lock down the traditionally open-minded LPG software the way you locked down TIGCC.

WTF?! I actually opened up TIGCC, it was me who created the CVS repository! Before, patches were exchanged by private e-mail and only releases were public. ld-tigcc wasn't even publicly available at all before TIGCC 0.95 Beta 1. For 0.96, I did all the development in the open.

If for you, "openness" means accepting any and all patches, then I have sad news for you: no project is "open" in that way. Even Wikipedia reverts changes which are obviously incorrect.
This was totally unacceptable for the sake of the community - and an utter disrespect for Julien and Romain, who have spent lots of time and energy into raising those projects to what they are. They didn't want to see their effort misappropriated by a well-known sabotager.

Then they surely shouldn't have given you access. roll

After sabotaging TIGCC, you did the same to the LPG and you dare blaming me for it. bang
LOL, I was sure that you were going to blurt out that obscenity rotfl

rotfl The truth is now obscene? laught
avatar
Mes news pour calculatrices TI: Ti-Gen
Mes projets PC pour calculatrices TI: TIGCC, CalcForge (CalcForgeLP, Emu-TIGCC)
Mes chans IRC: #tigcc et #inspired sur irc.freequest.net (UTF-8)

Liberté, Égalité, Fraternité

15

Ce sujet a été coupé en 2 afin de séparer la discussion principale des posts ./6 ./7 ./9 ./10 ./11 ./12 ./13 ./14 . Pour vous rendre sur le nouveau sujet, merci de cliquer sur ce lien
avatar
Ben, bouh, quoi :D

16

Please stop trolling here, and go on the other topic, thanks
[Folco -> balise foireuse]