 |
BorlandTalk.com
Borland discussion newsgroups
|
| View previous topic :: View next topic |
| Author |
Message |
Kristofer Skaug
Guest
|
 Posted: Thu Aug 26, 2004 12:40 am Post subject: What Delphi users really want (other than bugfixes) |
 |
|
As follow-up to the thread below concerning D9 feature requests and
voting, I was curious what sort of issues would be likely to dominate a
"feature-voting fest", so I took a statistical round trip of the QC
Delphi area (non-.NET) and recorded the following statistics:
For categories E and F (enhancements/new feature requests), Delphi for
Win32 has a total of (hold on to your hats) *855* non-closed reports, of
which 175 are Open (the rest are "Reported").
Out of these, 33 are marked "duplicate".
That means we have 822 distinct feature requests that are open for
voting/rating.
Now I checked the breakdown over areas within Delphi:
The most popular categories are:
IDE - 296
Compiler - 198
VCL - 103
Database - 53
Midas - 38
RTL - 37
Debugger - 32
-----------------------------
Total: 757
-----------------------------
That leaves some remaining 65 reports (less than 10%) for all the other
categories together, including such gung-ho stuff as Bold, AddOn, CORBA,
Enterprise, Internet, MTS and SOAP.
What does this mean? That these features are just so perfect that nobody
can think of anything to improve them?
Or does it mean that the vast majority of Delphi/Win32 users simply
don't care very much about these features?
Anyway, the numbers are clear: There are almost 300 feature requests on
the IDE itself, 200 for the Compiler and 100 for the VCL. A clear sign
that these aspects of Delphi have received rather too little attention
lately.
Looking at the current "Top votes" within categories E and F, we ignore
for a moment the "off the charts" items because they are either off
limits for D9 or they are not really features as such that need
developing (the dbExpress issues).
7324 - Create native 64-bit compiler/IDE - 453 votes
7837- Integrate Delphi and BCB - 128 votes
4007 - source code for all dbExpress drivers - 122 votes
3732 - dbExpress updates to be available separately from IDE/VCL
updates - 76
That leaves us with the following "top voted" ranking among the "new
features" as per today:
679 - The infamous WITH resolution - 58 [Compiler]
2747 - Undo/redo in form designer - 51 [IDE]
3427 - Resourcestring return WideString value. - 26 [Compiler]
376 - Save and select different project options for a project - 25 [IDE]
1230 - Provide translated Win32 API headers with Delphi - 25 [RTL]
8752 - When a Panel is created, the caption should be empty - 19 [VCL]
8513 - Extend CDS.RefreshRecord to include nested datasets - 19 [Midas]
2152 - Bundle DUNIT with Delphi - 18 [IDE]
7258 - Fastcode Functions in the RTL - 15 [RTL]
4530 - Support for code alignment through ...ALIGN keyword - 15
[Compiler]
3040 - dbExpress error codes - 15 [Database]
3166 - Delphi should support UNICODE - 15 [International]
2010 - Make Delphi interfaces as 'pure contracts' - 13 [Compiler]
1881 - CheckBox in DBGrid for boolean fields - 13 [Database]
1138 - Register custom debug-"renderers" for types of variables - 13
[Debugger]
649 - User defined Folders in Project Manager - 12 [IDE]
(due to the hop-skip query sequences needed to compile this list, I may
have missed some toward the bottom).
.... and then there are hundreds more with fewer or no votes.
FWIW,
--
Kristofer
|
|
| Back to top |
|
 |
Kristofer Skaug
Guest
|
| Posted: Thu Aug 26, 2004 1:00 am Post subject: Re: What Delphi users really want (other than bugfixes) |
|
|
Dave Nottage [TeamB] wrote:
| Quote: |
Just remember that voting is only part of the process. Sometimes even
the highest voted items might not be possible for all kinds of
reasons,
|
This in particular is why I "truncated" the top 4 items, which are not
only popular out of proportion, but are also "off the scope" for exactly
such reasons. And yes, we all know that QC voting isn't everything, but
it's the only thing we've got, so we better use it, right? <g>
--
Kristofer
|
|
| Back to top |
|
|
Will DeWitt Jr.
Guest
|
| Posted: Thu Aug 26, 2004 9:07 am Post subject: Re: What Delphi users really want (other than bugfixes) |
|
|
Kristofer Skaug wrote:
| Quote: | Looking at the current "Top votes" within categories E and F, we
ignore for a moment the "off the charts" items because they are
either off limits for D9 or they are not really features as such that
need developing (the dbExpress issues).
|
Yes, but isn't part of the issue? That Borland appears to be ignoring
issues that are clearly very important to those who use QC? Ignoring
the highest voted items is tantamount to silencing hundreds of people
and focusing instead on the smaller/smallest groups/issues.
| Quote: | 7324 - Create native 64-bit compiler/IDE - 453 votes
|
Obviously I can't be objective with regard to this report, but really,
the longer Borland puts this off the more native developers they stand
to lose. (I know I've already begun a new project using Visual C++
2003 (with the promise of being able to recompile it with Visual C++
2005 to target AMD64 *and* IA64), and I'd really rather not continue to
go down that road).
Danny spoke of "pain points" I believe in one of his blog entries--
well I can imagine no pain bigger than being forced to move to another
product because the one you've used for over a decade simply isn't
going where you need it.
| Quote: | 7837- Integrate Delphi and BCB - 128 votes
|
This sounds distinctly similar to the rumors about Diamondback (Delphi
and C# in the same IDE). Why would this be something that's "off
limits"?
| Quote: | 4007 - source code for all dbExpress drivers - 122 votes
3732 - dbExpress updates to be available separately from IDE/VCL
updates - 76
|
These last two seem to involve Borland having a change of vision of how
they plan to deal with customers with regard to dbExpress. For as many
votes as they get, that something hasn't changed (that I'm aware of,
I'm not a dbExpress user) already is a bit telling.
Kudos for getting the top voted/rated items into an easily digestable
format, I just take issue with downplaying the highest vote getters for
questionable reasons.
Will
--
Want a 64-bit Delphi compiler for AMD64 / IA-32e? Vote here--
http://qc.borland.com/wc/wc.exe/details?reportid=7324
|
|
| Back to top |
|
|
Dave Nottage [TeamB]
Guest
|
| Posted: Thu Aug 26, 2004 9:47 am Post subject: Re: What Delphi users really want (other than bugfixes) |
|
|
Will DeWitt Jr. wrote:
| Quote: | I just take issue with downplaying the highest vote getters for
questionable reasons
|
What are the questionable reasons?
--
Dave Nottage [TeamB]
See Delphi 9 (codename: Diamondback) at BorCon (US) 2004:
http://bdn.borland.com/article/0,1410,32499,00.html
|
|
| Back to top |
|
|
Kristofer Skaug
Guest
|
| Posted: Thu Aug 26, 2004 10:12 am Post subject: Re: What Delphi users really want (other than bugfixes) |
|
|
Will DeWitt Jr. wrote:
| Quote: |
Yes, but isn't part of the issue? That Borland appears to be ignoring
issues that are clearly very important to those who use QC? Ignoring
the highest voted items is tantamount to silencing hundreds of people
and focusing instead on the smaller/smallest groups/issues.
|
I sympathise with your concern, but the point I was trying to make here
was that these items (integrate Delphi/BCB, Native 64-bit Delphi) are not
merely "new features" but rather changes on an architectural/fundamental
product level, which need to be something that Borland wants to do
themselves and target a new release, or make a new product (like D8.NET)
specifically for this purpose.
IOW I think these are not "feature requests" but "new product requests",
and compete on a completely unequal footing with the smaller scope
feature requests that would fit into the existing product. As such, they
are not interesting in the scope of my list.
Regardless how many votes you get for 64-bit Delphi, it's clearly not
going to happen for D9 (or Borland would've announced this long ago), so
it's rather a waste of votes when Borland may currently be evaluating
"any other feature request except that one". I have personally shifted my
own votes from "I-know-this-aint-gonna-happen-for-D9" reports to more
realistic near term targets for the specific purpose of jockeying
*feasible* D9 features.
--
Kristofer
|
|
| Back to top |
|
|
damian marquez
Guest
|
| Posted: Thu Aug 26, 2004 10:20 am Post subject: Re: What Delphi users really want (other than bugfixes) |
|
|
Kristofer Skaug <ya.ierfgnf (AT) thnxf (DOT) x> wrote:
| Quote: | As follow-up to the thread below concerning D9 feature requests and
voting, I was curious what sort of issues would be likely to dominate
a "feature-voting fest", so I took a statistical round trip of the QC
Delphi area (non-.NET) and recorded the following statistics:
|
Best post in months 
Thanks Kristofer, really INTERESTING information....
|
|
| Back to top |
|
|
Frank de Groot
Guest
|
| Posted: Thu Aug 26, 2004 12:50 pm Post subject: Re: What Delphi users really want (other than bugfixes) |
|
|
"Kristofer Skaug" <ya.ierfgnf (AT) thnxf (DOT) x> wrote
| Quote: | "any other feature request except that one". I have personally shifted my
own votes from "I-know-this-aint-gonna-happen-for-D9" reports to more
realistic near term targets for the specific purpose of jockeying
*feasible* D9 features.
|
Then how will Borland ever implement 64-bit compilation?
If we don't vote for it, it will never happen (= Delphi's death virtually
guaranteed).
|
|
| Back to top |
|
|
Kristofer Skaug
Guest
|
| Posted: Thu Aug 26, 2004 1:24 pm Post subject: Re: What Delphi users really want (other than bugfixes) |
|
|
Frank de Groot wrote:
| Quote: |
Then how will Borland ever implement 64-bit compilation?
If we don't vote for it, it will never happen (= Delphi's death
virtually guaranteed).
|
I'm not saying you shouldn't vote for it, I'm just saying you needn't
vote for it *now*, because it's very likely out of the question for D9
anyway.
Instead, you can use your precious Delphi votes *now* to push less
ambitious (but still very useful) features of your liking into D9. Once
D9 is out (and perhaps your favorite features have become reality), you
can shift your votes back over to the long-term wishes (or the new D9
bugs, whichever is more important to you at that time <g>).
BTW I don't agree that 64-bit Delphi will never happen unless you vote
for it. Borland aren't quite as stupid or market-blind as some people
appear to think. If/when a 64-bit compiler becomes essential for Delphi's
survival, they will surely produce it.
It is far more important IMO to use votes for the less-obvious things,
those small but very lovable IDE features (such as e.g. code browsing),
which make you wonder how you ever got along without it. These are not
subject to business strategy, but are still very important in terms of a
happy end user.
--
Kristofer
|
|
| Back to top |
|
|
Dennis Landi
Guest
|
| Posted: Thu Aug 26, 2004 2:48 pm Post subject: Re: What Delphi users really want (other than bugfixes) |
|
|
"Frank de Groot" <franciad (AT) online (DOT) no> wrote in message
| Quote: | Then how will Borland ever implement 64-bit compilation?
If we don't vote for it, it will never happen (= Delphi's death virtually
guaranteed).
|
We did vote for it; and nothing happened. I think this issue in a nutshell
simply taught developers to ignore QC. The fact the so many votes still
accrue to the 64-bit tells me that the previous participants have simply
moved on to activities with a little more inherent integrity and potential
for reward on time spent....
-d
|
|
| Back to top |
|
|
Frank de Groot
Guest
|
| Posted: Thu Aug 26, 2004 3:06 pm Post subject: Re: What Delphi users really want (other than bugfixes) |
|
|
"Dennis Landi" <nada (AT) nada (DOT) com> wrote
| Quote: | We did vote for it; and nothing happened. I think this issue in a
nutshell
simply taught developers to ignore QC. The fact the so many votes still
accrue to the 64-bit tells me that the previous participants have simply
moved on to activities with a little more inherent integrity and potential
for reward on time spent....
|
Are you referring to Free Pascal and Lazarus?
I see that they already support 64-bits compilation and a lot of Delphi code
can be compiled.
Any idea when it will become a serious contender (when will it be able to
compile just about everything of the popular 3rd party components like
VirtualTreeview, Toolbar200 etc.)?
It would be great if Borland would get a competitor in Free Pascal, perhaps
then they will start listening to Delphi users - or have they given up on
the product alltogether?
|
|
| Back to top |
|
|
Kristofer Skaug
Guest
|
| Posted: Thu Aug 26, 2004 3:19 pm Post subject: Re: What Delphi users really want (other than bugfixes) |
|
|
Dennis Landi wrote:
| Quote: | We did vote for it; and nothing happened. I think this issue in a
nutshell simply taught developers to ignore QC.
|
QC is easily one of the best bug reporting systems I've seen.
As such it demands respect, and I would prefer it anytime over other
known "issue tracking systems".
(remember Borland's old support interface? Write a report, get an
automatic e-mail confirmation, and that was it).
The problem is not QC but the way Borland is handling it.
Obviously they cannot respond adequately to technically honour 800
feature requests for a single product, let alone all the bug reports.
And clearly they cannot let themselves be cajoled into dozens of
man-years of work in all urgency, just because a hundred or so Delphi
users would really like to have a 64-bit compiler yesterday. But I'm sure
they got the message and will consider 64-bit very seriously after D9.
What Borland could do much better at, however, is finding ways to reward
participation in QC, to keep the flame burning, to let us know they
appreciate the effort we invest in improving *their* product, offer some
direct feedback (in QC) on our ideas/reports, tell us when they've fixed
a bug so we don't need to wonder and worry about it needlessly for
another 3-5 months before the next release ships, set up short term
"pick-your-favorite" vote-fests for minor things which they commit to
implement, award extra vote quota's for people who show significant
volunteer spirit, awards for detecting duplicate reports, best and worst
submission of the week/month contests, etc.
Also they could do us the favour of vetoing/cancelling/deleting reports
that are absolute no-go's from the start, to save the reviewers the
trouble of reading and rating it too, and to spare the proponents the
agony of false expectations.
--
Kristofer
|
|
| Back to top |
|
|
Kevin
Guest
|
| Posted: Thu Aug 26, 2004 3:22 pm Post subject: Re: What Delphi users really want (other than bugfixes) |
|
|
Dennis Landi wrote:
| Quote: | We did vote for it; and nothing happened. I think this issue in a nutshell
simply taught developers to ignore QC. The fact the so many votes still
accrue to the 64-bit tells me that the previous participants have simply
moved on to activities with a little more inherent integrity and potential
for reward on time spent....
|
I must jump in here since I haven't really expressed my view in the past
because I didn't want to dampen the efforts of the "we want 64-bit
Delphi crowd" (of which I actually am a member believe it or not but I'm
also a realist).
I must say that I view a 64-bit Delphi compiler in the same light as
Kylix. It is nice-to-have but will Borland really make enough money
from their effort? How many people will _REALLY_ use this? You're
talking about a niche of a niche.
The only way we'll really know is to see what actually happens regarding
..NET and Win64 in the long-run. If .NET is an abysmal failure in the
64-bit world then Win64 may stand a chance. However, I can't imagine
most people choosing a native 64-bit compiler over a .NET compiler when
it comes to the 64-bit platform. The majority of people will choose
..NET. Why? Because you can run the same binary on both platforms.
When you have to start building two different binaries you're starting
to get into deployment issues and maintenance issues. Most people would
rather avoid this. Plus, once you're talking 64-bit with .NET you're
also opening up doors to other platforms through Mono. For me this is
the one area where .NET really does start to make sense (AFAIK, the only
other one is web apps or other server-only apps). This will be the key
driver for me in picking .NET to do my development in the future: to
support 32-bit and 64-bit processors equally well with the least effort.
Before that time I do not need .NET. And that time is rapidly
approaching.  Delphi 9 is probably in that timeframe.
BTW, I still view Kylix as important and a little more important than a
native 64-bit compiler. It just needs to work a bit better: the whole
deployment thing is an issue. I really hope that Mono quickly becomes a
standard part of the major distributions and then Kylix can evolve into
a Mono dev tool rather. However, that is quite long-term and I wish
Borland could do something about Kylix a bit sooner. A little service
pack and QT3 support should do nicely. Its just that I understand
Borland's position- Win32 and .NET are the bigger fish right now.
I'm not saying that people shouldn't vote for a 64-bit compiler. It's
just that I hope Borland makes an intelligent decision about it and
doesn't let other important developments suffer as a result.
Cheers,
Kevin.
|
|
| Back to top |
|
|
Craig Stuntz [TeamB]
Guest
|
| Posted: Thu Aug 26, 2004 3:28 pm Post subject: Re: What Delphi users really want (other than bugfixes) |
|
|
Dennis Landi wrote:
| Quote: | We did vote for it; and nothing happened. I think this issue in a
nutshell simply taught developers to ignore QC.
|
Um, what do you expect the turnaround time for Borland to support a
totally new platform should be from the time when people first express
interest? Consider the time it took Borland to support Linux and .NET
in formulating your answer. Consider also that two years ago this
newsgroup was choked with paens for Borland to deliver real support for
..NET.
I think -- and this is speculation, mind you -- that it is not the
case that "nothing happened." I think that Borland is doing work that
they're not talking about, which may or may not turn into a production
compiler for Win64.
I can understand wanting fast turnaround for defects in QC, but new
platform support is going to take time to do well.
-Craig
--
Craig Stuntz [TeamB] . Vertex Systems Corp. . Columbus, OH
Delphi/InterBase Weblog : http://blogs.teamb.com/craigstuntz
Please read and follow Borland's rules for the user of their
news server: http://info.borland.com/newsgroups/guide.html
|
|
| Back to top |
|
|
Frank de Groot
Guest
|
| Posted: Thu Aug 26, 2004 3:36 pm Post subject: Re: What Delphi users really want (other than bugfixes) |
|
|
"Kristofer Skaug" <ya.ierfgnf (AT) thnxf (DOT) x> wrote
And clearly they cannot let themselves be cajoled into dozens of
| Quote: | man-years of work in all urgency, just because a hundred or so Delphi
users would really like to have a 64-bit compiler yesterday.
|
Dozens of man-years?
Plural of dozens is a multiple of 12.
It should not take a single person more than a year, to make 64-bit
compilatiom, and that is a very liberal timeframe.
|
|
| Back to top |
|
|
Frank de Groot
Guest
|
| Posted: Thu Aug 26, 2004 3:37 pm Post subject: Re: What Delphi users really want (other than bugfixes) |
|
|
"Kevin" <kevinbe71 (AT) y_a_WHO (DOT) c_o_m> wrote
| Quote: | I must say that I view a 64-bit Delphi compiler in the same light as
Kylix. It is nice-to-have but will Borland really make enough money from
their effort? How many people will _REALLY_ use this? You're talking
about a niche of a niche.
|
Look at it this way.
How many people still use 16-bit compilers?
Many shops here MAINLY sell 64-bit machines nowadays.
How long will a 32-bit compiler sell AT ALL?
|
|
| Back to top |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
