A cross-platform vision for Delphi

[Sascha Willems]

If you follow this link (http://delphiroadmap.untergrund.net/A_cross-platform_vision_for_Delphi/index.html) you'll find a big proposal from Simon Kissel (the man behind CrossKylix) concerning a cross-platform vision for Delphi which basically says that Borland should stop focusing on .NET and should revive Kylix to also serve the Linux userbase.
He already posted an open letter in 2006 (link (http://delphiroadmap.untergrund.net/)) with an alternative roadmap for Borland that would also include linux as a target. He has gotten a lot of feedback (even from Borland) and recently also contacted the newly formed CodeGear. But sadly he had gotten not a single answer from them so he decided to make this whole thing public.
So if you're also interested in seeing Kylix again (hopefully at a lower pricetag or even together with a free Turbo Express edition) head over there.

[AthenaOfDelphi]

Thanks for that Sascha. Quite an interesting read. Lets hope that CodeGear actually listen to their user base and restart the project.

[WILL]

I think from what Simon is saying here is that he believes that he hasn't been listened to. So he kind of gave up on it and released all his material on the talks after only 3 weeks.

If you ask me; personally, I feel that he gave up too soon. After only 3 weeks? Big companies, especially after just transitioning from a major reformation of the much bloated Borland, takes time. I would have waited at least a couple of months before I called off the whole thing and threw in the towel.

I very seriously doubt that anyone from CodeGear will take him up on this or any future proposals after seeing how quickly and easily he folds under the least bit of resistance. I think he screwed the pooch on whatever leeway he had made in getting them to listen to his ideas. Now he just seems like a hot-headed Delphi fan who didn't get his way.

A company like CodeGear doesn't need this. And hastiness will only bring more problems to the already daunting task that this company has before it as it stands. Thats my take on this situation.

[jdarling]

I completely agree WILL. While he makes good points, and ones that MANY of us have stated, his impatience with the process places him in a bad view. Corporations take time, 3 weeks in a corporation is about 3 hours in a normal persons life.

It is not uncommon to see change proposals take months to go up and down the food chain. They must go through legal review, and replies must be properly formulated. In a small business or personal world this isn't the case. Our replies can be swift and curt. In a corporate world the reply must be thought through and the source looked at closely. A single person within the corporation cannot make the reply or provide information to the submitter without first having the details of the reply reviewed. Otherwise the corporation can be sued by the submitter.

Adding in a section at the end "What I can do for you" is also not a proper thing to do when you are; attempting to address the issues presented by the user group at large. If Borland liked his ideas, and thought he had a point, then they would have asked him to spearhead the movement. By telling Borland what he could do, he has essentially placed himself in a position of competition.

On the topic of CodeGear stating that it would not pickup Kylix again; This does not mean that they will not do a cross platform version of Delphi. It means that they will not be basing any future work on Kylix, nor do they have plans on utilizing that particular project/product name again.

His final salt to the wound was placing it on Borland's NG. There is no worse assault that you can take then to attack someone on their ground. Of course, Borland can't remove the post, this would make them look even worse. They have to reply, carefully, and that in turn leaves a bad taste in their mouths.

In short, I wish the author had taken the time, proper channels, and approach. Instead, he might have doomed an otherwise seemingly common and valuable point of view.