Re: 100% pure Win32 IDE

lurkio wrote:
Okay, if we agree on the necessity of the "relatively low cost" part
then I am with you :-)
[...]

I'm glad we're making progress. ;)

Not in theory; developers are using it and for ASP.NET applications,
AFAIK, it works.

The problem for me is that it only ever brings (and is always fated to
ever bring) a functionally incomplete re-heat of an old version of .NET
to *nux...and the worry for me in that context is that once you attract
people into getting a taste of .NET, they will eventually want (or even
need) to use the real thing and you'll have given them the perfect
platform to switch away and never come back to you. Tim Anderson's blog
had a prime example of this as few months back :

http://www.itwriting.com/blog/?postid=401

IMO, this doesn't matter as long as BDS can sell into those
organizations. ;)

Regardless, it appears to be a problem with mono's garbage collector.
This is where DevCo can put pressure to dedicate an engineer to help fix
the problem. Certainly this will require more investment, but overall,
it's less expensive than building your own runtime. ;)

BTW, this is a mirror of the kind of thing that is happening to Delphi -
there is a clutch of folk who stuck with Delphi.NET for a little while but
then left for VS to get their hands on generics, et cetera - and I'm pretty
sure that more will be leaving to get their hands on LINQ, etc. months
before it appears in Delphi (even with deference to the potential time lag
reduction in the new DevCo era)...

Sure, which is why IMO, C# support in BDS has to be intensely
competitive in a head-to-head way.

Just to clarify, supporting Mono also applies to *nix only development;
hence it's not always/only about cross platform development.

In the context of *nix only development, why should Mono be
coupled to MS technology /at all/ ?

Otherwise it would be Mozilla. ;)

Why can't they get on with
forging ahead in those areas where MS isn't going instead ?

They can, but traditionally, Delphi has been a Windows tool.

If people need what is offered from .NET, they generally get
it straight from the source...

"generally" is the key word there. As mentioned many times on this NG,
all DevCo needs to do is be profitable. They can do so by leveraging a
competitive advantage within a certain niche.

But isn't there a glaring irony here if that's how folk
see Mono ? If it is being mooted as a potential escape
route from a putative MS lock-in, why is Mono technically
based upon, erm, mimicking an MS technology at every step
of the way (apart from the bits they can't keep up with)
and therefore is ultimately tied at the most fundamental
level to the .NET roadmap ? Now, that's what I would call
/serious/ lock-in :-)

Not at all. The easiest way to migrate away from something is to move
to something 100% compatible and have the option to go back.

Although SQL is a horrible at being compatible (and many including EF
Codd are/were highly critical of it), there's a reason why almost all
DB's support it. DB's of course could come up with their own query
language and refuse to play IBM's game by mimicking them. ;)

Miguel de Icaza has a history of bringing MS tech/ideas to Linux; you'll
have to ask him about his motivation.

Well, it goes back to that cost thing - if it doesn't cost much
for DevCo to support it and they get some ROI on it then I can
agree with you on them doing something about it.

I believe they can and I also believe a smart investment in supporting
mono will improve their MS .NET ROI as well.

--
Brian Moelk
Brain Endeavor LLC
bmoelk@xxxxxxxxxxxxxxxxxxxxxxxxxxxx
.

Relevant Pages

  • Re: Codegear in the news
    ... Like BDS2006, once you go past the rudimentary support Mono offers for .NET, it becomes near useless as a programming tool/platform. ... About as Rudimentary as Lazarus' support for Win32 VCL development (and people are producing commercial applications with Lazarus, ... And, of course, Mono is harder to deploy on Linux than the .NET Framework on Windows, due to dependencies. ... The same fears that everyone has IRT Microsoft are real on Linux, a platform known for its ability to break existing applications on upgrade, and known for the plethora of patches that can come through via automatic updates. ...
    (borland.public.delphi.non-technical)
  • Re: comparisons of open source .net implementations
    ... Mono is the forerunner. ... For GUI stuff mono will support GTK#, Tickle#, Winforms, ... winforms once fully supported (check mono's website for more info as to ... do some gui stuff and I want to know which project enjoys better support ...
    (microsoft.public.dotnet.general)
  • Re: 100% pure Win32 IDE
    ... Mono is what it is; it provides value to those that use it. ... of how good mono actually is and I doubt MS will ever support it. ... BDS can and should go farther than merely marketing ... If you want *nix support and like .NET, ...
    (borland.public.delphi.non-technical)
  • Re: 100% pure Win32 IDE
    ... there will be space for things like ECO and third-party ... Mac support is another example. ... ..NET that makes feature lag in both Mono endemic. ... Releasing a new COBOL compiler "would be something different ...
    (borland.public.delphi.non-technical)
  • Re: Remobjects as a role model
    ... CodeGear was DevCo for quite some time, and everyone at 'DevCo' was expounding on their newfound freedoms from Borland management. ... One of the liabilities was the need to support a product that needed a lot of support. ... The marketing still isn't done well, and there is so much scrambling to support a lot of BONEHEAD decisions left in place (and in many places strongly supported by current management) that I don't know that it will ever change. ...
    (borland.public.delphi.non-technical)