D7 revival? stirring the old pot + Delphi 64 suggestion

I'll stir the old pot: how about a D7 revival?
The new compiler, but with the old IDE, the old help. Nothing fancy: just get the old source, replace with new RTL/VCL/Compiler and get it to run. No need for the new code editor or whatever extras, if it needs .Net, or if it needs too much conversion work, don't include it. Period. Keep things simple and stable.

Just thinking of that after seeing some polls on french developper site "developpez.net":
http://developpez.net/forums/showthread.php?t=24265
D7 and previous still account for 60% of the user base, and you can see by the previous polls that things are barely changing.
There is a an idea: sell upgrades to what Delphi users use. Keep them in the fold rather than let this user base erode away...

As for D.Net, the figures speak of themselves:
http://developpez.net/forums/showthread.php?t=24266
You can check other areas of the site and poster stats, it's not that Delphi posters don't use .Net, it's just that they don't use Delphi for .Net work.

Additionnally, and this is the beef of this post, I would suggest considering releasing *very* *early* versions of the Delphi 64 compiler and RTL, even with an IDE as bare as it get (a very crippled D7-IDE would do f.i. ;)), under a license that'll allow fastcode challenges to take place for key areas of the RTL (the MM, math stuff, string stuff, the core classes...).
As for the 64bit debugging, a CPU view debugger would be all that is really needed to get things started (this is for fastcode after all), no need for rich source-level debugging, inspectors, etc. early on.

I would insist on the *very early* nature of the release so as to make possible challenges on core classes such as TList, TStrings, TInterfaceObject (constructor/destructor internals) or the various internals of string manipulation (including UTF strings, as I hope there will be).
This wouldn't need to be a free-for-all mess, CodeGear could issue "grants" for specific areas (in the form of points f.i.) to orient things.
Validation could be two-ways: one in the fastcode challenge, another made via CodeGear private unit tests (then just give the result: pass or fail).

I would also insist on this being an open (non private) release, so that public challenges can take place (private challenges aren't much challenges), and the outcome is less susceptible to individual's free time.

This would accomplish two things:
- send out the clearest message there could be that CodeGear is serious on 64bit (which isn't clear right now), and keep the current native Delphi users around rather than looking for a way out.
- get a better 64bit compiler & RTL out of the box

Hopefully it could free up some CodeGear manpower by reducing the optimization and unit testing workload (CodeGear would be free to reintegrate fastcode unit tests into their main test suite), though more realistically, this freed up time would be spent as community relations.

Eric

PS: for whoever is in charge of the D2007 trial "downloadability", I would suggest adding a torrent of the complete trial download. The current 2007 trial link goes to a Java applet which - here - just sits there doing nothing.
.

Relevant Pages

  • Re: D7 revival? stirring the old pot + Delphi 64 suggestion
    ... D7 and previous still account for 60% of the user base, ... sell upgrades to what Delphi users use. ... really needed to get things started (this is for fastcode after all), ... via CodeGear private unit tests ...
    (borland.public.delphi.non-technical)
  • Re: Codegear in the news
    ... that's the best way for the existing product line to improve the way we want and improve support for what /many/ developper are asking: .net 2,3 etc, 64 bit compiler, compiler enhancements, improved VCL etc... ... Not all Delphi developers are Delphi only developers. ... If CodeGear works in the PHP/ruby etc arena I hope they won't have a dependency on .Net et al as I would want to run it under Wine on Linux. ...
    (borland.public.delphi.non-technical)
  • Re: Interface for Compiler Godegeneration Optimization Modules
    ... CodeGear do not have resources to add optimization modules to the ... then be able to help CodeGear improve the Delphi codegen. ... compiler part of the w32 delphi compiler they don't have the resources for, ... Currently i never debug apps with optimization on btw, ...
    (borland.public.delphi.language.basm)
  • Re: Registration sucks again
    ... - Automated builds that use the command line compiler, ... Backups of virtual machines with activated copies of Delphi. ... CodeGear RAD Studio). ... IMHO, the best thing they can do is remove the need to activate from older versions of Delphi (for example by making a "final version" available for download, which requires a key, but no activation). ...
    (borland.public.delphi.non-technical)
  • Re: Delphi for mac?
    ... I hope we will have Delphi on Linux too. ... With Kylix, Borland made the mistake of listening to their ... this compiler instead of trying to keep up with Microsoft with the .net ... Guess that (unless you're the owner of CodeGear) it's not your call. ...
    (borland.public.delphi.non-technical)