Hello David,
> Recovery is very possible.
I just wanted to make sure of this.
> However, with the current leadership and product line, I for one,
> am not willing to make a financial bet in the outcome.
And I understand and respect this.
> OK, on technology, and I will only comment on the product line
> today.
> 1) JAVA. What in the heck are you trying to accomplish with it?
I think that this is one of the most interesting thing in the industry today. No one, even Sun, has come out and stated that we are all trying to "out abstract" Microsoft! That's what this whole thing is about. All of the companies working on Java are simply trying to make Windows a trivial component in the computing model. If Java succeeds, then you will be buying Java apps. And running them on any Java platform (software or hardware) from any supporting vendor. And of course Windows might be that platform, but it might not be!
> Of course you web server needs to have a JVM. All mainstream
> operating systems have that today to be competitive.
The Web server doesn't need a JVM to serve Java applets to browsers ... only to host Servlets for doing Web Server based applications or extensions and processing. Yes the Novell Web Server needs to have Servlet support.
> Are you trying to offer a faster, better java client for the PC?
This is actually a difficult question. I've sat here and written an answer three times and still can't think of how to explain this clearly. I guess the answer is "yes" ... but not in a traditional sense. Java creates a whole new model of computing (that I believe Microsoft is going after with COM, DCOM, ActiveX, and Windows) that really blurs the lines of "client" and "server" and starts to create migratory objects that execute, in a distributed fashion, all over the network. Where ever it makes sense. Is Novell working on technologies to enable this? Yes. Are we writing a traditional "Java client" ... I don't think so.
> Where will you MAKE MONEY WITH JAVA?
Another good question ... it's like making money with the Web. It's the indirect dollars in many ways ... not the direct dollars. I believe we will make money on many of our products because of the services that we offer to support Java environments.
> 2. INTEL-only. Of course it has the highest number of seats, and
> some day it may catch up to RISC.
Catch up in what way? Are you suggesting a Beta vs. VHS argument here? Intel dominates, and Andy will not have it any other way. Is it the "best" technology? Well ... again another question. ;-)
> (If SUN, IBM, DEC, etc.. fire their R&D departments :)
>
> The point is, that if you insist on INTEL-ONLY for your Border
> packages & groupware, then the RISC-vendors treat you as
> competition. Port your code to Alpha, PA-RISC, SPARC, PowerPC, etc,
> and see the big boys become your friend, not enemy.
The big boys? Is Intel not a "big boy"?
I would suggest that every attempt to move in that direction has led to failure because the volume for these processors (or machines based on them) does not exist. Look at Microsoft and NT ... if your statement was true, then why have they bailed out of every processor but the Alpha?
And for a moment think about why the Alpha? I believe that Microsoft uses this as a serious tool. The Alpha provides an engineering tool to verify that NT stays cross platform ... just in case. But I also believe that the Alpha is used to keep Andy in check. Bill always has his own little hardware "stick" to smack Intel with from time to time.
Nope. I don't buy it. It's not worth the effort today for such a small verticle market. Major customers have no interest in any platform but a standard Intel motherboard.
> We commented on your hunch? that the telcos will implement
> bordermanager as "caching technology"???
There is no hunch. I firmly believe that as more and more people understand the power of hierarchical caching they will implement cache servers. I also know that BorderManager implements this technology and will be one of the (if not *THE*) first commercial product offering this power. And I'm not picking sides when I say that the telcos *will* implement technologies such as this.
> Not while thier friendly IBM RS6000, and HP9000 reps are telling
> them that they shouldn't trust this problem to be solved with
> "personal computers running Novell software, or microsoft software,
> either".
Well ... I think the tides are changing. It's one thing to say that Novell won't get in, but are you also saying that Microsoft won't get in? They have recently made a series of announcements of implementations of SS7 and TMN/CMIP in NT ... I think Bill's planning a frontal assualt! And Andy and Intel will be assisting in the charge!
> Honestly, would you trust, say, a nuclear power plant to be
> controlled and monitored by Intel-based personal computers?
Of course. Now what software would be running on it? NT? ... which seems to blue screen every couple of days? Of course not. But I'm lost at what a RISC chip, CISC chip, Intel Chip, HP Chip, IBM Chip has to do with it. Software is what crashes computers ... poorly written software. I just read about the plane crash on Guam. It appears that a software bug prevented the ground personel from detecting the improper flight path ... I'll be it wasn't running on a PC ... it was probably running on one of those reliable RISC boxes? ;-(
> Same thing. PC stands for personal computer. This stigma, fair or
> unfair, is going to be around for a long time.
I agree with this ... but that seems to be what Bill is up to with his "Scalability Day" events, etc.
I'd argue the tide is rapidly changing ... the lines are blurring ...
Scott C. Lemon |
