Howdy All,
On 30/10/2004, at 3:14 AM, Andrus Adamchik wrote:
> If you haven't seen my posts on this same topic in the past, they are
> in
> the archives. So I won't start with "I love WebObjects..." part this
> time,
> and go straight to the "...but" part ;-)
Understood, and the same for me.
> In the argument "WO vs. the Rest of the World", WebObjects is still
> winning on its technical merits (though this is almost not noticeable
> now
> compared to say 1999).
I'll come back to this in a second.
> But when you deal with real-world situations, there
> are other factors that count. One of them is perceptions and prevailing
> mindset of the people you work with (be it your clients, investors,
> supervisors or employees).
I agree with this too.
Now coming back to "WO vs. the Rest of the World." I don't think I am
really debating anything with Andrus (for I think he was just making a
passing reference to this). However, I would like to make a point (or
two).
I think the claim that "WO is still winning on its technical merits
(and for me development paradigm) but that this is almost not
noticeable now, compared to say 1999" is probably true *just* for
standard object-relational mapping and Web-control frameworks.
However, there are a couple of things that WO still leads significantly
in:
1. Client-side editing contexts (although probably not as "fully
functional" as they could be). I'm not sure of any other O-R mapping
tool that provides these (although my understanding and hope is that
Cayenne is implementing these in a future version).
2. Direct-To-Web/JavaClient/WebServices. The whole rule-based
development of the presentation layer (and beyond) no matter whether it
is a HTML interface, Java interface or Web services interface. Apple
has taken these ideas from "research" to "production."
I believe the first will become critical when people move to include
thick clients with their HTML clients (who really wants to use a HTML
client all day long).
I believe the second (D2*) is just way ahead of its time (like WO was
back in 1995+). The rest of the community will get there, perhaps,
sometime in the next 10 years.
> So while one can argue that NSArray is superior to java.util.Collection
> and friends, NSArray still remains one of those things that scare the
> hell
> out of an average Java developer (and maybe not so average too).
Just for interest, and I realise you were only using this as an
example, but may I ask why?
1. Because it is different
2. Because it separates NSMutableArray, NSArray?
3. Something else
4. All of the above.
> And there
> are other things too...
>
> And I am not even going to start on the Apple's refusal to market/sell
> WebObjects as an enterprise Java solution directly competing with
> weblogics and jbosses of this world. This alone made a dozen or so
> companies that I know to switch away from WO completely.
Yep, that's the only real reason WO is not going forward. (I believe)
Apple has decided it no longer wants to be in the enterprise
application development tools and platform market (boy that's a mouth
full). It's got iPods to sell (and problems with declining Mac market
share).
That's unfortunate, but probably a good business decision.
> On the other hand with Cayenne I rarely (if ever) see any of these
> concerns. So to me it has always been a pragmatic solution to Java
> persistence problems instead of an uphill battle for a technology
> religion.
Yep, it's a strange world (of politics and biases).
Just my 2 cents too.
Cheers,
Ashley.
-- Ashley Aitken Perth, Western Australia mrhatken at mac dot com
This archive was generated by hypermail 2.0.0 : Sat Oct 30 2004 - 05:48:45 EDT