Re: Fwd: Intel native MCL

As the person who designed and implemented a lot of the MCL IDE, and
who has contributed to the CCL IDE, it's really quite simple.  A lot
more time was taken on the MCL IDE than the CCL IDE.  That's all there
is to it.  MCL was a commercial product, so we could pay people to
work on it. CCL is open source and not much money has been invested in
the CCL IDE.  There haven't been a lot of code contributions either.

There's nothing fundamentally wrong with the CCL IDE.  It just needs a
lot of TLC and attention to detail to make it as good as MCL.  In
fact, we (Clozure) consider Hemlock to be a sounder base technology
than Fred.  We carefully weighed porting Fred versus continuing to use
Hemlock as the base, and we decided to stick with Hemlock because it
has a lot more EMACS functionality than Fred ever did, e.g. the
ability to have keyboard macros.  Fred is better written (gz wrote it
and there's no one better), but Hemlock has more functionality.

Unfortunately MCL will die with Lion.  I find it incredible that it
has survived almost 25 years with happy, active users the whole time!

There are three things needed to make the CCL IDE great:  an overall
design, testing, and attention to detail.  We need an overall design
to make sure that it's not just a collection of random menus, dialogs,
and features.  Then it's just a matter implementing features and
finding and fixing bugs.  Probably the best way to maintain UI
coherence is to have a benevolent dictator like Linus or Guido, but
that will probably take money.

So if you want the CCL IDE to be as good as MCL: contribute money, or
code, or testing, and at the very least submit Trac tickets for bugs
or desired features.



On Mon, Jun 6, 2011 at 3:24 AM, Benjamin Perrault
<[email protected]> wrote:
> Agreed. Broken in many a facet.
>
> This is just an example of why MCLIDE and MCL "feel" right.
>
> They don't have ( or as many at least ) wonky behaviors, which furthers my concern that these will go away when Lion comes around.
>
> The fact is that MCL or MCLIDE ontop of Clozure CL is a substantially better experience and I have to wonder if, though perhaps the technical merits of migrating things to the intel platform may not be there, the end user experience is worth the pain. Frankly, I don't want to end up back in Emacs + SLIME world - which isn't nearly as nice a place to work as FRED and the general MCL / MCLIDE environment.
>
>
> On Jun 5, 2011, at 11:55 PM, Toomas Altosaar wrote:
>
>> Figured out what was happening but not what went wrong:
>>
>> My repeated evaluations were actually being processed by Clozure CL but the Listener window was not scrolling to the correct location. It just sat at the same location giving no indication that new lines were being sent to it.
>>
>> I realized this when calling up the backtrace window and realizing that I was in an error situation. I also realized that I could scroll the Listener downwards for pages (via mouse scrolling) ... while the buffer location indicator didn't move. Once at the bottom it seems like he Hemlock buffer is now in synchronization with reality.
>>
>> It is just these types of things that scares me enough to stay away.
>>
>> Basic broken functionality.
>>
>>
>> 10 >
>>> Error: value NIL is not of the expected type NUMBER.
>>> While executing: CCL::+-2, in process Listener(6).
>>> Type cmd-. to abort, cmd-\ for a list of available restarts.
>>> Type :? for other options.
>> 11 >
>>> Error: value NIL is not of the expected type NUMBER.
>>> While executing: CCL::+-2, in process Listener(6).
>>> Type cmd-. to abort, cmd-\ for a list of available restarts.
>>> Type :? for other options.
>> 12 >
>>> Error: value NIL is not of the expected type NUMBER.
>>> While executing: CCL::+-2, in process Listener(6).
>>> Type cmd-. to abort, cmd-\ for a list of available restarts.
>>> Type :? for other options.
>> 13 >
>> _______________________________________________
>> info-mcl mailing list
>> [email protected]
>> http://clozure.com/mailman/listinfo/info-mcl
>
> _______________________________________________
> info-mcl mailing list
> [email protected]
> http://clozure.com/mailman/listinfo/info-mcl
>
_______________________________________________
info-mcl mailing list
[email protected]
http://clozure.com/mailman/listinfo/info-mcl