In this second and final part of my interview with Ted Shelton, Borland’s chief strategy officer, Shelton focuses on Microsoft .Net’s support for multiple languages and reveals a hidden .Net stumbling block. First part here.
Borland’s Bird’s Eye View
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
Despite .NET or java or whathever being better. Most developers just work to get paid, most use windows, and most use whathever microsoft puts in front of them (visual studio in this case). On these issues companies can choose between opposing microsoft (SUN) and not following it in the tech market, or just follow microsoft and eat the crumbs is MS lefts. We know that whathever company does anything good, microsoft will try to buy that company, and if it cant, it will try to copy it.
All in all, in this article borland simply says to their actual and possible future customers that they are still loyal to microsoft, so it’s safe to keep buying borland stuff because it wont go “outdated”. Period.
Does anyone thing that ‘web services’ are going to go the way of channels and push technology? Perhaps I’m just informed, but ‘web services’ seems to be more like the darling child of somebody’s marketing department than something that’s actually useful for developers and end users.
Is the end result of all this to try and get everyone running their applications over the web? And if so, why? Who besides the corporations are going to benefit from this kind of model? I can see it now … I’m working in a web-based word processor, and then my Internet connection goes down, and now I ca’t get work done. Either that, or the word processing company decides I need to spend $500 to upgrade to version 12 (just one month after buying version 11), and then I don’t have access to the old version anymore because word processing company is pulling all the strings and removed version 11 from their web services menu
Isn’t the main use of web software the same as it has always been for
the client-server model – access to a single database for workers who
are spread around over a large building, or on the move with portable
computers?
That could include simple word processing for entering a report into
the database, but the sensible thing in most cases is to prepare
anything larger than a few paragraphs in a local program, then import
the result to the database management program.
At one time it made sense to store big programs (such as CAD) on a
sever and download to client computers for running, perhaps even using
overlays. That approach is obsolete now we have 40 Gig drives.
Networks are slow, so they should only be used when there is no
faster alternative. The Internet is very very slow.
I’m not sure I buy into the webservices thing completely yet.. With each additional integrated (into my web apps) webservice lies the possibility for yet another connection failure point (try telling your clients it’s not YOUR fault an app/feature is unavailable). It’s tricky enough troubleshooting a remote DB connection.
On the other hand .NET is an exciting idea especially for us Windows developers who are looking to (in the near future) port applications to Linux.
Cheers,
E.
will borland’s .NET solution also work under Linux? that would be a great market for them to get into, a little dangerous considering they need to keep a good relationship with Microsoft, but i think it would be a worthwhile risk for them
will borland’s .NET solution also work under Linux? that would be a great market for them to get into, a little dangerous considering they need to keep a good relationship with Microsoft, but i think it would be a worthwhile risk for them
Worrying that your chief competitor (MS & VisStudio) might not like a strategy (eg, moving into Linux) is a poor reason to avoid a market.
The “power” of Web Services, really, is their “light weight” and, ideally, commodity status.
Light weight is somewhat of a misnomer today, SOAP is anything but Simple, but that was one of the original forces behind WS.
Commodity is that fact that they can be really really stupid. They don’t need a lot of the overhead that CORBA layers on top of them. By being “simple”, then everyone “gets them for free”.
Every major programming and scripting environment today has relatively easy access to web services, usually via a library or a module.
Since they’re easy to implement on both the service and client end, and most programmers have easy access to them, then they can become more ubiquitous.
So, that means that if you’re creating some kind of server protocol for Something, you may as well use some kind of web service framework rather than not. Unless your looking at a zillion messages, the overhead isn’t worth the headache of contriving your own protocol. The web service frameworks give you a format “for free”, and you get to focus on getting the data moved.
So, now, it is getting safer and safer to make the assumption that anyone wanting to connect to a custom service, that it will be using one of these web service frameworks.
The big question is whether web services will take over from EDI. Something will.