.Net Basic Questions

[minoad]

I have 2 Quick questions....

First off, has anyone seen an IDE published by any company besides Microsoft for .NET?

Second, has anyone tried to "Integrate" ASP 3.0 applications with ASPdotNet, and if so could you tell me how it went and if you had any difficulties and what they were? I am the responsible developer for a rather large site and would like to begin adding .Net pages (I think I would be crazy not to) but at the moment do not have the manpower to convert a couple hundred files over.
I appreciate any help.

Micah A. Norman
ShopCrc.com

 

[link9]

Macromedia has a new one coming. They're actually re-vamping their entire suite, and it's named MX.

DreamweaverMX integrates the functionality of DW and UltraDev 4 into one program, and I saw it demo'd at a conference this past week. I also have my hands on the preview release (which you can download at their website). I think I still feel the same way about it that I did with the other, though. VID for development, and DW for design. They do have code support (homesite integrated, as well), but it's just not as good as VID, especially now. VID has gone through some amazing changes, and is nothing short of spectacular now, which is alot more than you used to be able to say about it.

I saw someone else using something that I **think** had 'slick' in it, like 'slickedit' or something, but it was sort of like noteTab or the like -- glorified notepad, so you don't gain the advantages of having a true .net IDE, like you would with MX or VID. More are supposed to be on the way though. The framework is very extensible, and that leaves the door wide open for many companies to make their own IDE's with their own plug-ins and custom extensions.

As far as integration... there really isn't a good migration strategy. Here's the thing.

Most of your ASP classic code will work with .NET. However, it won't take advantage of the new framework, and therefore will not give you the performance that you'd expect with .NET.

One example is early and late binding variables. In ASP classic, everything was late-bound (variant data type). In .NET, you have to (well, don't HAVE to, but you do by default -- you can turn it off, but that wouldn't be smart) declare all variables and instantiate them to a particular type... early binding.

In performance tests (presented by MS, so take it or leave it), MS took an existing ASP classic page, and changed the extension to aspx. They then made any required modifications to make the code work... tested it, and the code ran a little faster (5% or something).

The next thing they did was go through and early bind all variables used. That's it. Just typed everything. That was reported to increase performance by over 100%.

The more things you do to move code over (ADO to ADO.NET for instance), the more performance gains you'll see. Problem is that it's a HUGE undertaking to do things like that. ADO.NET for instance, does not have the recordset anymore. Now I don't know about your code, but mine pretty much relies on recordsets. That would mean a complete and total re-write of all pages using recordsets. Don't hold your breath on me doing that.

Instead, I'm going to move forward from this point in pure .NET code. Old systems will stay like they are (I might go through and early bind them, but that would be the extent of it), and new systems will come online with .NET.

In one session I attended, a program manager from MS stated "there isn't a good migration strategy". Now that's from the horse's mouth, and not a public announcement or anything. That's just a personal opinion from someone on the .NET team. They wouldn't make that statement publicly in fear of running developers off, but the fact of the matter is that this new platform is so radically different from the old, that taking existing pages and converting them just doesn't make good sense.

Instead, take the essence of existing pages, and just re-write them taking advantage of things like output caching, and re-useable user controls and custom web controls. I bet if you really mapped out a strategy, you could do it in alot less time than you think.

I'm lucky, as most all of my sites have a life-cycle of between 6mos - 1yr. Reporting sites that go the way of the dinosaur once everyone's seen results. So I don't have this problem to address (at least not much).

Anyway, hope this isn't too discouraging. It's just my honest opinion, and it's worth exactly what you paid for it. ;-)

good luck!
paul