Xref: utzoo comp.sys.apollo:5697 comp.sys.hp:5565 Path: utzoo!attcan!uunet!wuarchive!mit-eddie!uw-beaver!ubc-cs!alberta!arcsun.arc.ab.ca!calgary!cpsc!dan From: dan@cpsc.ucalgary.ca (Daniel Freedman) Newsgroups: comp.sys.apollo,comp.sys.hp Subject: Re: Netpower: encourage HP to improve customer services Keywords: HP Apollo security patches APRs bugs customer services net Message-ID: <1990Jul6.155846.7327@calgary.uucp> Date: 6 Jul 90 15:58:46 GMT References: <1990Jul5.142403.3942@quintro.uucp> <1990Jul6.015817.23710@metro.ucc.su.OZ.AU> Sender: news@calgary.uucp (Network News Manager) Reply-To: dan@cs-sun-fsd.UUCP (Daniel Freedman) Organization: U. of Calgary, Calgary, Alberta, Canada Lines: 93 In article <1990Jul6.015817.23710@metro.ucc.su.OZ.AU> jimr@maths.su.oz.au (Jim Richardson) writes: >In a number of recent articles in comp.sys.apollo, people have expressed their >dissatisfaction with many aspects of HP's customer services, in particular >with HP's attitude to security. > >For example, in <1407@m1.cs.man.ac.uk>, root@craven.ee.man.ac.uk (Operator) >"Colin" writes: > >>Frankly, it seems to me that Apollo's attitude towards security sucks, which >>is a great shame 'cos I generally love the machines. > >I couldn't agree more with Colin. With a real networked file system (//), >token ring, the DM, ACLs and more, Apollos have a system which could be >driving most other Unix workstations out of the market (no offence to HP >machines intended :-) But the company's services to customers are bad enough >to cancel all this out. > [...much deleted...] >What can we do? Unfortunately this has been going on for years, and every few months, someone gets really frustrated, and posts a message such as yours. Trust me, Apollo doesn't seem to care. As you said, a few guys obviously do care, but they have not been able to change things all that much. Lets ask a question: Why would anyone in the world buy an Apollo workstation today? 1) Price? Well, since you can buy a faster sparcstation SLC for only a little more than Apollos cheapest machine (the 2500), price does not cut it. 2) Performance? The 10000 is pretty fast, but also pretty expensive. Also, we have heard that the 10000 is more or less obsolete, to be replaced with a combined HP/Apollo risc machine. MIPS has faster machines, and a Sparc 490 is not to be sneezed at for similar or less dollars. 3) Technical Superiority? It is true that Apollos have a technically better O/S. The distributed object oriented file system is - if slow - at least innovative and functional. However, (maybe people will want to comment on this) I feel that although some features are missing or brain-damaged on Suns, you can get anything to work on a Sun that you can get to work on an Apollo. In other words, Apollo doesn't offer anything that you fundamentally can't get on a Sun. Since Suns are more open than Apollos, and since there is *much* more Sun expertise out there than Apollo expertise, the "hassle differential" between making something work on a Sun and making it work on an Apollo is greatly reduced. Technical superiority of the o/s, while an important factor in the past, is now largely irrelevant. 4) Support? I think Colin's and Jim's messages tell you all that you want to know about this one. 5) Direction? If it looked like Apollo was going to come out with some really neat stuff, then this might be a reason for not migrating away from Apollos -- for perhaps at least deferring the decision as to what to buy. However, from what I've seen, Apollo is going to continue to do in 1990 and 1991 what it did in 1987-1990 (and possibly before that too). That is, it announces hardware for 6 months from now which has slightly better performance than today's machines from other vendors. Of course, 6 months from now, the competition will be selling machines which are twice as fast, for a similar price. 6) No choice? Well, if you have made a large investment in software which uses proprietary stuff from Apollo such as d3m (the database), then you are probably stuck with it. This is probably the only good reason to buy an Apollo. However, at some point you are going to want to bite the bullet and move to something which doesnt tie you down, such as oracle. The fact that oracle may be inferior is irrelevent. You can do everything that you can do with d3m with oracle (although it may be distasteful), and once you've done it, you can move it around at will. In short, its not that Apollo has changed, its that it hasn't changed while those around it catch up. The point has been reached where other vendors machines allow you to do just as much, quicker, cheaper, and in some cases due to inreased availability of tools and expertise, better. I think it is really sad that Apollo has frittered away all of its (substantial) advantages by simply *letting* the competition overtake it. If it sounds like I'm frustrated, its because I am. I have been a faithful fan of Apollos for 4 years, having administered a netwrok of 15 Apollos in a predominantly SUN department. It used to be that I could get away with saying "these machines are better and neater than the Suns", and it would be true. Today however, the answer from our Sun users is "so what, I can do what I need to on a Sun, and it runs n times as fast". Dan Freedman U. of Calgary Computer Science Dept., 403 220-7299 2500 University Dr. N.W., dan@cpsc.ucalgary.ca Calgary, Alberta, Canada. T2N 1N4