My Proposal for a project (from Alvar, http://alvar.a-blast.org/)

Create and support a project, which delivers an easy installable IDE for Perl, which can be used by persons who are no CLI freaks (new users!), with debugger, editor with syntax check while typing, support of Perl::Critic in GUI, support of perltidy within the GUI, with lots of essential modules, mod_perl, Catalyst, Jifty, Gantry, good tutorial in more languages than englich (german, french, spanish, ...), ...
The package should be ready for use for the most common OS: Windows, OS X, Linux; FreeBSD, Solaris, ... -- the X11 systems are more or less the same ;-)

And the project website should be placed good visible on the usual Perl websites, with fast download links (like Firefox: http://www.mozilla-europe.org/de/products/firefox/)

After that, create press releases ("business", "professional" and other bullshit bingo words).

You say: too much work?
No, at least for the beginning: take Eclipse and E.P.I.C (http://e-p-i-c.sourceforge.net/), use some not so ugly colors ;) and create bundles for each OS with this - including Perl, Modules and perhaps some other related Eclipse plugins (SQL, HTML, ...)

You say: vim is much more better?
Maybe better for you. But such developers are in danger of extinction.

contributed by guest@hidden on Dec 1 5:46am

Whatever you may say about IDEs - there are many programmers out there who won't even think about using a particular programming language if there's no IDE for it.
Judging about a language based on the tools available for it may not be fair but there is some sense in it, because they do save time especially for the not-so-experienced users.
So I strongly second Alvars suggestion.
If we want Perl to attract more attention then a good cross plattform IDE would be a major step in the right direction.
Perhaps it needn't be an IDE for Perl 5.x (there's EPIC and Komodo, which basically work) but having a fully functional (syntax highlighting and checker, debugger, code completion, etc.) IDE (based, e.g., on Eclipse Dynamic Languages Toolkit) ready for Parrot/Perl 6 would be sufficient.

Just my 2 cents

contributed by guest@hidden on Dec 2 3:48am

Please do not create a new IDE, if you want there to be good IDEs for Perl, please improve the ones that already exist. It would probably do more good for more people to improve the debugger and other tools then to start a new project.

contributed by guest@hidden on Dec 10 10:49pm

I know about only two sensible cross platform IDEs for Perl: EPIC and Komodo.
Last one is closed source.
First one is free but - as I understand - heavily tied to Perl 5 and not using newer developments for Eclipse like DLTK.
I like the idea of having sth. ready for Perl 6 / Parrot when - or before ;) it´s finally there.
Perhaps a general debugging interface for Parrot is possible which could be reused for a Perl 6 IDE?
May be bullsh*t - don´t know enough about Parrot internals...

contributed by guest@hidden on Dec 11 11:09am

Might be worth chatting with Andy about the work he did for his refactoring engine grant http://www.perlfoundation.org/march_25_2007_adam_kennedy_s_refactoring_editor_grant

contributed by Adrian Howard on Dec 13 5:46am

Basic rules to success!

• Improve any IDE you like adding support and special features focused on Perl
• Configurable to work with any key-binding (vim, emacs, eclipse, etc)
• Fast, easy and extensible

contributed by guest@hidden on Dec 21 12:16am

I'm very pleased to find your comment, since it's an issue overdue to be addressed. It's not that good Perl programmers need a real IDE -- I'm quite happy using mostly Emacs, and I like the ability to modify the Perl mode (with lisp) as desired. But Perl6 is a different matter -- a lot of new constructs to learn, and a lack of general information about syntax and behavior. Even just the discipline of compiling a "help" menu for a Perl6 ide would help centralizing this information -- web searches alone bring up contradictory summaries of many new features, reflecting different stages in the evolution of Perl6 or different writer's opinions on what would be in the final specs. Developing code completion and syntax highlighting tools would be a great way to learn the langauge inside out -- and someone actually doing that work would have a good rationale to write to the big fellas to ask how something works.

I'd also like to opine that the initial use of Perl6 may largely be as a kind of "paradevelopment" language, and those of us who believe in it should promote its usefulness in that area. What I mean by "paradevelopment" is that even a project whose actual shipped, or open-sourced, code is in some other language (C++, C#, Lisp, whatever) could use Perl6 for things like testing, versioning, ensuring stlyistic and policy-driven consistency across a compilation unit (or project or whatever), and code-generation. Perl6 grammars are wonderful, and they seem to look and act like a cross between regular expressions and parsing expression grammars (I noticed that the author of the new "spirit" parsing library on boost cited Perl6 as an inspiration). Someone needs to prosletyze them more in computer science circles, 'cause the world still hasn't caught on to how great even Perl5 regexes are (entire algorithms can be embedded in regexes with lots of $+s and ?({})s). I happened to learn a little about the regex 'make' function at Frozen Perl '09 and it has only gradually dawned on me how theoretically interesting and subtle it is -- in effect, it embeds a concept of types directly in the regex system, implying a kind of typed Kleene algebra. Anyhow, Perl6 makes it easy to define grammers for any language and therefore analyze source code, so it can be a great tool even when the primary development occurs in some other language ("progamming with Perl6 is not necessarily programming in Perl6"). This should help jump-start acceptance of it, because it can be used strickly backstage -- but we need to develop modules which promote this usage.

Anyhow, a Perl6 IDE could exemplify that concept even if it is written, say, in C++; Perl6 could be used to test and analyze the C++ code as it is under development. I admit to being skeptical about a good IDE written only in Perl; MOST IDEs written in the same language as they target don't work well. (I have Java IDEs written in Java that are so slow as to be worthless.) Of course, since parrot would have to be embedded for debugging (all the existing Perl IDEs seem to run scripts by forking an external process -- is that really necessary?), users could write extension scrips (like Emacs macros) in Perl (if one of the parrot lisps get done maybe we could support porting emacs settings also and emacs-like key sequences). I'd love to work on such as IDE and, if folks disagree about these suggestions/ideas, perhaps we could work independently but touch base over time and potentially merge at some point. Meanwhile, there are several people whose suggestions I'd like to court re Perl6 IDEs and related topics, so perhaps I'll try to gather some other opinions and post them as I get them. Thanks for initiating this needed discussion!

contributed by guest@hidden on Mar 14 8:31am