David asked: > Some of these problems can probably be put down to the user > being used to doing things one way and frustrated at having > to do them another way (particularly if the new way isn't > easy to work out). However some of the problems sound like > they could be more serious in that the new functionality is > noticeably worse than it was in Delphi 5-7. > > I gathered from the below post that Accredo have moved to > Delphi 2006. I imagine there are some other reasonable sized > development teams which have moved up to Delphi 2005-2007. > > So... how have Accredo and other similar development teams > found it? How many gotcha's really are annoying rather than > just a change in the way things work?
> My interest is obviously that we are still using Delphi 5 > (despite owning various newer copies as well). I'm still > planning on moving forward when Delphi 2008 comes out but I > am a little concerned... Firstly, we essentially upgraded for the IDE improvements and BDS 2006 was initially unusable due to not being able to handle projects the size of Accredo so it wasn't until Update 2 (or later - I can't remember exactly) that it became viable. They obviously had some nasty O(n squared) algorithms in there (where n is number of project elements). Also, remember that we use all our own GUI controls and TDataset layers, so most of the various VCL gremlins that others are finding don't affect us. And that's one of the reasons we use our own stuff - quality of implementation is in our hands, we're not relying on anyone else not to let us down. To be fair, we did have to do some component work initially to resolve a couple of low level VCL changes and to work with the IDE package cacheing and deferred loading but that wasn't too painful once we got to the bottom of it. > It sounds like the Help system making gigantic leaps back > between Delphi 7 and 2005+ is one annoying gotcha. What else? Well, firstly I'm still mainly working in Delphi 7 since I'm maintaining the current release codelines, whilst the other members of the team are working using BDS 2006 on our next release development codelines but with the odd bit of pair programming I've done - the new IDE has some nice features and improvements in the debugger which I consider very valuable. (Sidenote: IMO, debugging is harder than coding, so if you write overly 'clever' code that is at the limit of your coding ability or comprehension ability, you may find it outside your ability to debug!) But yes, I do feel the Delphi help has been limping along getting very little love from Borland/Inprise/Codegear for years, BDS 2006 is just the latest iteration of this trend. And that's largely a management priority/resource allocation issue. Historically, I don't think Borland has put much importance on the Delphi help. But that's been true of Delphi as a whole product too - what resources Delphi has been allocated when into the features, not the documentation, which is a common oversight in todays marketing driven world. New features on the tick list produce more sales (generally) , you only find out what the product is really like (documentation, etc) after purchase. It's called post-purchase evaluation in the traditional marketing model of the customer buying process. So, speaking with my cynical marketing hat on, features are more important than documentation if you want sales now and are thinking short term since you're not sure you want to invest more in the product's future. Now, to be fair that's more the old Borland, whereas CodeGear do seem to be trying to change this (finally!) but it takes quite a bit of time and effort to execute on a new vision and turn any organisation bigger than 2 people around... Time will tell... TTFN, Paul. _______________________________________________ NZ Borland Developers Group - Delphi mailing list Post: [email protected] Admin: http://delphi.org.nz/mailman/listinfo/delphi Unsubscribe: send an email to [EMAIL PROTECTED] with Subject: unsubscribe
