Page 2 of 8

Re: CompilerDev?

Posted: Wed Sep 25, 2013 10:46 am
by d2alphame
I wish there was another wiki/forum like this for writing/developing programming languages (compilers/assemblers/interpreters/virtual machines)... Something like forum.compilerdev.org and wiki.compilerdev.org

Re: CompilerDev?

Posted: Wed Sep 25, 2013 12:13 pm
by dozniak
It's not easy to set up a community, but if chase and mods are not against it, we could have a corner of this forum dedicated to that.

Re: CompilerDev?

Posted: Thu Sep 26, 2013 4:43 am
by d2alphame
That would be sooooo totally cool!

Re: CompilerDev?

Posted: Thu Sep 26, 2013 6:04 am
by fronty
Or everyone interested in compiler development could just use comp.compilers.

Re: CompilerDev?

Posted: Thu Sep 26, 2013 2:20 pm
by ~
fronty wrote:Or everyone interested in compiler development could just use comp.compilers.
The same could be said for any other topic (including comp.os.*).

Sometimes I don't understand why people complain about diluting a community towards external resources, and why sometimes they insist in adding resources to the community, or even relying on education from external resources. I guess that within any community there are subgroups of people with both views, and not everyone can be made happy at all times, and even then what already exists could be enough to bring projects and discussions that add to any topic, including compiler development, among so many others, for the ones interested, as has always been.

But anyway, for anyone interested in going to Usenet, you can use Thunderbird Portable for that and a tutorial to set up an email account for your installation (preferably GMail or other one that is easier to set up than Yahoo, and follow something like this tutorial) and adding Usenet to that account and then you can use a public Usenet server like nntp.aioe.org.

Re: CompilerDev?

Posted: Fri Sep 27, 2013 12:11 am
by dozniak
~ wrote:But anyway, for anyone interested in going to Usenet, you can use Thunderbird Portable for that and a tutorial to set up an email account for your installation (preferably GMail or other one that is easier to set up than Yahoo, and follow something like this tutorial) and adding Usenet to that account and then you can use a public Usenet server like nntp.aioe.org.
Thanks, that's my main gripe with USENET - the need for additional software and setup. As opposed to this forum, which is available everywhere.

Re: CompilerDev?

Posted: Fri Sep 27, 2013 6:19 am
by bwat
Are we going to have a subforum here (I think there's lots of overlap between programming language implementation and OS implementation --- it's all good old fashioned systems programming after all), or will we start a new forum somewhere else for language implementation (I may be able to host if nobody else wants to)?

Re: CompilerDev?

Posted: Fri Sep 27, 2013 9:11 am
by fronty
~ wrote:The same could be said for any other topic (including comp.os.*).
Not exactly. Saying that about for example operating system development would be like saying "you should abandon your working established community and join some other community". In my opinion that's radically different to saying "you shouldn't try to forcefully create a new community instead of joining old, well established and respected community".

Re: CompilerDev?

Posted: Fri Sep 27, 2013 9:42 am
by Antti
If we randomly picked up a topic from the "comp.compilers" and from the "OSDev.org compiler subforum", it would be more likely that the latter one is more interesting to me. Here we could concentrate on topics that can accompany our OSDev hobby. I do not want to underestimate current compiler communities. However, perhaps there are too much diversity because the "compilers" topic is quite broad.

Re: CompilerDev?

Posted: Fri Sep 27, 2013 10:38 am
by ~
fronty wrote:
~ wrote:The same could be said for any other topic (including comp.os.*).
Not exactly. Saying that about for example operating system development would be like saying "you should abandon your working established community and join some other community". In my opinion that's radically different to saying "you shouldn't try to forcefully create a new community instead of joining old, well established and respected community".
After all, the quality and effectiveness of the material of a community so that beginners can progress steadily depends on the capabilities of every individual contributing to that community and, more importantly, the amount of knowledge that is actually shared in an accessible and complete way.

Yes, it seems that improving the quality of the questions and sometimes the answers, or at least increasing the attempts to post threads about generic and reusable information of all kinds in a rich and practical way should be enough as a new starting point to better satisfy the technical needs that are left behind in the most publicly accessible parts of communities like this.

There are plenty of options and indeed they don't involve forcing things. It's fine if no new communities are created, since there is the "General Programming" section, topics that fit perfectly in other sections, and a lot of other places to learn from, as well as a few books and tutorials.

It's just that in the end, regardless of the topics, what would be nice would be to discuss and document more of the useful things (for the users) that can be done with an OS. I see that without such discussions, which is already difficult enough everywhere, there is usually no room to properly talk about a lot of things involved, stay on topic, and have people actually contribute to a discussion. I see that only a very few OS projects have become advanced here, but for the rest things seem to have become stalled (the fact that almost anything that has been implemented in such projects has been discussed clearly here or somewhere else could be regarded as a warning that something is increasingly lacking). So it would seem that so many resources wouldn't make so much sense after all as would have been intended for the majority if there is no proper discussion/disclosure of topics beyond the basics, to make more unusual things considerably more familiar. And if the projects get abandoned and no knowledge is documented for all of the tiny advanced bits, there is really not very much benefit after the projects and its developers disappear.

There must be some other ways to encourage talking about advanced topics in an effective way that could be understood gradually, with heavy help, be it direct or from documentation, because right now it's all about finding and understanding the more advanced things 100% virtually by oneself.

There is the Wiki, this forum, the IRC chat, there are books, tutorials, Usenet groups, sites like Stack Overflow, but what is lacking is the practical exposure to explanations that stem from the real world, what has already been done, big and small, and that take them apart more clearly. Regular asking/answering almost never solves it properly because of poor questions or too specific to a particular implementation/problem/bug. It seems to happen everywhere that there is a discussion/questions board.

The only thing that occurs to me for that, but not necessarily the best that there could be done, is to define a practical goal/purpose, take a little piece (for a demonstration program), and explain it so it is complete in itself, and taken step by step initially ignoring everything that doesn't apply for that particular purpose (instead of relying on lots of resources and full, raw standards and specifications, having everyone find out the same things from scratch and not documenting at least a clear enough general progression of what it took for it to be done).

Re: CompilerDev?

Posted: Fri Sep 27, 2013 10:55 am
by Brynet-Inc
dozniak wrote:Thanks, that's my main gripe with USENET - the need for additional software and setup. As opposed to this forum, which is available everywhere.
Google has a public archive for Usenet, and many ISP's have their own server.. it's not that hard to access, and here's a shocker, not all good things on the Internet are exclusive to your browser.

Re: CompilerDev?

Posted: Fri Sep 27, 2013 11:30 am
by dozniak
Brynet-Inc wrote:
dozniak wrote:Thanks, that's my main gripe with USENET - the need for additional software and setup. As opposed to this forum, which is available everywhere.
Google has a public archive for Usenet
Well, thanks, Google.

Re: CompilerDev?

Posted: Fri Sep 27, 2013 2:00 pm
by ~
Brynet-Inc wrote:
dozniak wrote:Thanks, that's my main gripe with USENET - the need for additional software and setup. As opposed to this forum, which is available everywhere.
Google has a public archive for Usenet, and many ISP's have their own server.. it's not that hard to access, and here's a shocker, not all good things on the Internet are exclusive to your browser.
There are so many options now that they would seem to suggest that the best option would be trying to get the best out of the different communities (this included), and sometimes ask/sometimes answer/sometimes show what one has done/sometimes attempt to explain what others have done in any of them which is most appropriate in a given moment.

Also, I remember that I was able to get things done faster when I just limited myself to read this forum and other resources, and it wasn't until I was able to explain some things and also when I needed information about poorly covered topics everywhere (at least publicly) that I started posting, but I figure that those things haven't been explained better anywhere because it would require a lot of additional work.

There seems to be a lot going on over IRC and Usenet, and probably other places, from OSDev people here, and still not enough explanations to do something new and outstanding from the public educative point of view, only non-generic issues too specific to a particular software for the most part. So a lot of new board topics everywhere are poorer than they should, and one would think that this is one of the main things all of this is intended to help at.

The result is that only a few can actually implement the advanced things in an useful way. I guess that solving all of it would involve a little more patience, less rushing, and trying to cover the things that are most interesting for each one in a more modular way.

Let's see what can be done from the basics. I personally like being able to stay in technical discussions, leaning and teaching, as good as I can, more than opinions. But in all existing boards, of all kinds, it is difficult to find the most appropriate times.

Re: CompilerDev?

Posted: Sat Sep 28, 2013 12:36 am
by bwat
~ wrote: There seems to be a lot going on over IRC and Usenet, and probably other places, from OSDev people here, and still not enough explanations to do something new and outstanding from the public educative point of view, only non-generic issues too specific to a particular software for the most part. So a lot of new board topics everywhere are poorer than they should, and one would think that this is one of the main things all of this is intended to help at.
I think it's the other way around - assuming I've understood you correctly. The generic stuff, w.r.t., compilers and OS stuff, has been done to death, but not on internet fora, instead it's in books. OS design, compiler design, and abstract machine design have been very well described by people who know how to teach. If you want book suggestions just ask.

What's missing, what's hard about it all, is the nitty gritty issues like:
  • How do I write a high level debugger for this language?
  • How do I write a control-C handler that throws an exception in this language so the user can break out of an infinite loop in his/her code and jump to their own handler?
  • How do I implement, test, and document a logarithm function?
  • How do I test my garbage collector?
  • How do I best let the user define their own syntax extensions to the language?
These things are still part of the black art of programming language implementation. The books that describe this are often old, obscure, and difficult for the hobbyist to get their hands on. A good forum will help with this.

W.r.t. comp.compilers, I've only had one conversation about that group with another professional compiler writer and we both had bad things to say about the moderator. It was quite funny actually.

Re: CompilerDev?

Posted: Sat Sep 28, 2013 10:04 am
by Combuster
dozniak wrote:It's not easy to set up a community, but if chase and mods are not against it, we could have a corner of this forum dedicated to that.
$.02 (and to get it back on track):

What I see with many people starting off on OS development, is that they can use a compiler from the instruction template they have always used - and for most desktop things you wouldn't need more than that. The result is that nobody knows how a toolchain really works until they get slapped hard by that mistake.
What happens next is that people will start to build runtime linkers - and at that point you will really need to know how the compiler outputs its stuff, and you will have to perform the necessary logic. Besides the task-specific linking, many system parts will end up having configuration files that will need to be read and parsed.
In other words, any sufficiently advanced OS has all the characteristic components of a (often simple) compiler, and depends quite a bit on the remaining parts of the compiler which the developer didn't need to write. In reverse, an actual compiler writer will need to be aware of how the OS deals with binaries without actually having to write that code.

So to include compiler development in this place would not be a far stretch and would probably have a symbiotic relation and a number of people migrating topics for fun and profit as a result.

----

Just as important is the follow-up question: would it take off if we tried? That one concerns me more actually.