Slashdot Mirror
Novell's Race Against Time
DiamondGeezer writes "The Guardian newspaper in the UK reports in 'It's a race against time' that Novell is on a knife-edge financially and competitively, having placed a huge one-way bet in the success of its Linux strategy. But there's no guarantee of success: its revenue from Linux licensing is puny, and it faces a crowded market of Linux distros. Novell may be getting some positive press now that it's gone full tilt for Linux, but let's remember the reasons why: because of mis-steps of its previous management (especially the disastrous acquisition of WordPerfect in the mid 1990s) and its failure to grow its Netware business (with more than a little help from Microsoft), it's now having to re-engineer itself for Linux."
Is Novell deserving of the support that the Open Source Software (OSS) community can provide to increase the odds of success regarding its Linux push?
If so, what can the average Linux user do to help (besides switching to Novell Linux Desktop (NLD) or becoming a shill?)
If not, why not?
How does Novell expect to remain competitive in the world of free linux. Especially with RedHat dominating the paid business sector.
They were one of the pioneers of many technologies available today. It will be sad to watch their slow painful death.
--
NoVA Underground: Where Northern Virginia comes out to play
Netware didn't just fail "with more than a little help from Microsoft". It failed because (and it kills me to say this), Windows NT was a better product than Netware in just about any way imaginable. I remember when I made the switch in my career from Netware to NT. I can't think of anything that Netware did better than Windows NT. Netware pretty much sucked ass...
I must admit that it was very early in my IT career that I made this switch. Perhaps my inexperience in Netware had something to do with my opinion of it.
Last post before the dreaded Slashdot April Fools articles are submitted.
I've been thinking for a while, from just before the start of the SCO vs IBM circus, that Novell is getting ripe for being bought out by IBM. Anyone else concurr?
The biggest problem Novell has is attracting the best and brightest software engineers. This is because a lot of Novell engineering is done in Provo Utah. Life in Provo is not for everyone. It is beautiful but it is one of those one company towns. If your job there does not work out then you'll have to relocate for your next job. The cost of living in Silicon Valley is high but a great engineer can find a new high paying job within a matter of days. Provo does not offer that.
While I can't argue much against that statement, I really wish it hadn't been disastrous. WordPerfect has always been my favorite word processing suite, and I wish Novell still owned it, and would give better Linux support than the wishy-washy stuff Corel's been doing.
Join moola.com, play games to earn money.
Novell may be facing competition with a lot of other distributions, but I have to say that I don't see Suse fading anytime soon. In fact, in my experience, Suse has been getting more popular as of late. It certainly seems like the most well refined distribution I've used lately. Redhat seems to have left a bad taste in the mouths of a lot of Linux users, and I've never heard of anyone using Mandrake on a server, which really leaves Suse as the last of the major distributions with commercial backing (I know there are other commercial Linux distributions, but when I think of commercial Linux distributions, I always think of the big 3 as Suse, Redhat and Mandrake).
YaST is probably one of the best system tools I've used on any Linux distribution, and hopefully we will see some really great things once we see some (forgive the buzz word) synergy between Suse and Ximian.
Famous Last Words: "hmm...wikipedia says it's edible"
I don't recall hearing that Novell ever intended to make a lot of money by selling a Linux distro. On the other hand, converting their existing products to Linux will save them a bundle compared to developing a completely independent solution. Most of what their stuff does is the same as what Linux does. They can concentrate on the things that make them different to add value.
I, for one, hope Novell makes a go of it, but the world is unfortunately a harsh place.
Engineering is the art of compromise.
Maybe Novell should also license from IBM all it can from Lotus SmartSuite, then merge the best of WP, Paradox, (if it still has any rights to WP & Pdx) into SmartSuite and then release the package to compete with OpenOffiice.org.
I am STILL not pleased with the document insert "feature" which, when I insert a document, it goes into a "band", invisibly. When I link to another document and then want to edit from beginning to end, I damn well should be able to SEEEEEEE those linked documents.
Lotus WordPro has done this RIGHT for years. 1.9.79 still makes me go, "sigh.... maybe NEXT version SO/OO.o will pull their heads out and actually BUY a copy of Lotus SmartSuite and start simplyfiying and mimicking stuff that works, instead of coming up with gee-whizz stuff that piles on to the list of features that have to be debugged, making it too resource intensive (on the devs AND the desktop) to be economical to go back and fix those useful features.
I also think Novell could pull a rabbit or two by adjusting the document interface so that a user in a spreadsheet can put the tabs wherever they want, not just get force-fed bottom tabs only or top tabs only. (Actually, I thought GroupWise offered that, or maybe it was QuattroPro...)
If IBM and Novell had some limited thing going on, Lotus SmartSuite could be diffused (not DEfused, mind you) across more Linux/Open Source environments.
I am not at all about "killing off" the various suites that OpenSource devs are making, but christ-o-matic, take a LOOK at what SmartSuite has, and gingerly, without ticking IBM off too terribly, clone some of those features, especially if IBM is not going to get its Lotus camp on the band wagon.
NOVELL, are you leesteneeng? Please, PLEASE, license from IBM/Lotus the Lotus and then uppgrade the Approach database interface, the Lotus WordPro document interface, and the Lotus 1-2-3 interface. They're crisp, tight, concise, colorful, not drab/gray.
At least Lotus isn't busy chomping away and cloning the heck out of ms' orifice. (Actually between Lotus and SourceNext ("Lotus SuperOffce developers/distributors in Japan/Asia...), I wonder what will be the next offerings to Lotus SmartSuite.)
David Syes
Previously: "Linux... Toward the Sunrise..." Now: "Linux... Toward the-- No, now, part of Every Sunrise"
What Novell really needs to do is merge unique features from Netware into Linux, and license much of Apple's proprietary code at any price.
Wow, sounds like a good idea. I'm sure Steve Jobs and Apple would jump at the chance to undercut their own market by allowing Novell to make a low cost Mac alternative!
We have a fairly sizeable eDirectory tree of about 100,000 users. We have hundreds of Netware servers. We use Identity Manager(dirxml) extensively. Our entire LDAP authentication runs on eDirectory. I know many other VERY large companies such as ours where Novell plays a very important role and where eDirectory is the central authentication/idenitity scheme. Sure we have some Windows application servers---who doesnt. But I always get amazed at those who predict the death of Novell---because usually those are the same people who have never used any Novell products in their life. Believe me Novell is dominate in every Fortune 100 company out there. They are going nowhere.
At the moment in my place of employment, we want to run Debian on some custom hardware (but alas, Debian won't work on it - despite many hack attempts), because we just find RPMs too hard to manage and apt-get + aptitude to ge great.
Employing someone to waste time trying to install Debian on something which cannot guarantee a pay off is not fun, and is a waste of money as well for the customer.
The thing is, the hardware vendor doesn't take Debian seriously (because it's not backed by a company with resources), so there is no driver disk or hardware support.
What I am trying to say is this:- there is a niche here that needs to be filled. There is a need for companies with the ability to back Linux distros, even if just for customer peace of mind. They will pay money for it. So far, only RedHat is being taken seriously commercially IMHO.
There is no "swamped" Linux market, or at least, not in my situation as far as I can tell. Only blimmin' RedHat Linux is supported and will install on our blade server.....
READY.
PRINT ""+-0
Novel vs Redhat. In a corporate environment with directory services novel wins hands down. Novel (suse) also has a much much better QA procedure on its Enterprise linux products. I have not seen to date one issue to date (yet) that has caused systems to go down after patches have been applied. However with redhat we see it all the time.
For anyone serious about an enterprise level linux novel is the only real choice.
While I don't doubt that Novell has taken a large risk with playing "the linux card", I don't see them in any immediate danger of financial difficulty.
http://finance.yahoo.com/q/is?s=NOVL&annual
While their installed base is certainly not what it once was, they have a solid reputation, still significant installed base, and from what I remember, a decent size pile of cash (771,844 at last quarterly report) to fall back on.
In other words, exactly where SCO might have been if they had not made a different sort of bet. (i.e. running a business of making products, selling support and consulting services, etc. Not to start an SCO love fest, but once upon a time they were a well regarded company).
Firstly, this fails to take into account the recent Microsoft settlement which brought $536 million into Novell, plus the additional cash they have on hand. They aren't suddenly going into bankruptcy.
Second:
"it is getting excited about the version of KDE that will accompany SuSE Linux 10 next year. This is based on Mono, another Novell takeover, which aims to provide a development environment that will run Java and Microsoft.net on Linux"
KDE has nothing to do with Mono. The author probably meant Ximian Gnome, but that doesn't even make the statement true, and wtf does Mono have to do with Java?
SuSE + Ximian + Mono + Novell = Good prospects in my book. Granted Novell management has a long history of screwing things up, but this product line looks pretty promising. In fact, full disclosure I put my money where my mouth is and purchased some amount of Novell stock.
The more you know, the less you understand.
I was thinking about this the other day after watching the BrainShare video.
All the stuff that will actually increase sales is based on Suse (clustering, Xen, etc...).
Why didn't they try to buy Trolltech instead of Ximian.
I just don't see how Mono is going to help the bottom line in the near term.
Heck, I would've have bought Trolltech, and slapped some proprietary apps into Suse. There's got to be a competitive advantage somewhere, and I don't know how just services is going to give them that.
Suse was already pretty much a KDE distro, and buying Trolltech would have given them two things.
(1) The ability to change the Qt license to a more liberal one.
(2) Bring in the talent of Trolltech that is already accustomed to working with Suse.
KDE/QT still has a superior framework to Gnome/Gtk+, but frankly I see the Qt license being the one showstopper that will push Gnome/Gtk+ into the "standard" desktop category, once Linux on the desktop actually matters.
My FreeBSD 4.x boxes are over 1000+ days of uptime now with real load with real business logic running on them.
ROFL. Yeah, right, sure. And rocks fall upward, and the moon is made of green cheese...
I administered Novell and NT both back in the days you're talking about. Netware 3.x beat NT 3.x so badly, on every possible applicable point, not even the most incompetent admin would have ever made the claim you just did with a straight face.
Microsoft PAYED my employer a huge sum to partially replace our Netware server with an NT 3.5 server. That little netware server was keeping the office served so far as email, web gateway, and file and print serving without a problem. NT on a slightly faster machine proved itself incapable of handling the EMAIL ALONE for the same office, and this AFTER having guys from MS fly down to work on it every couple of weeks for six months.
On top of lack of functionality and lack of stability, it was also impossible to properly admin. It was a total POS and everyone knew it. Even the PHBs were totally upfront about it - they knew it was trash. But whenever we had a problem, MS cut a check (or something to the same end effect) to more than cover the losses.
That's how they won. Novel certainly made mistakes, but that doesn't change the fact that their product was vastly superior and defeated on grounds other than technical.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Friends don't let friends enable ecmascript.
I would argue that Sun has done more work then any other organization with regards to OpenOffice, it doesn't seem to have translated into a large amount of cash. On the other hand, Novell does have a history of some great network management products, the Directory services spring to mind and I personally have never heard anything bad about ZenWorks. Red Hat would have nothing comparable if Novell were to whip them out and make them what I remember of them. I always liked the NDS.
"I use a Mac because I'm just better than you are."
I had a lot of problems just getting it installed.
perhaps its just because im not the type to use the 'gui' approach to things, not that its a bad thing.
it either didnt have drivers for hardware that we needed to function, or had quirks with those drivers, or needed special treatment.
HOWEVER, that being said i installed it on a plain machine... and the installation was flawless.
I dont like yast, it was good for getting things to work, but I ended up in a console with a text editor editing configs anyhow.
the cups and samba portions in yast are just completely barebones.
the built-in kde conguration tool is far more powerful with regards to samba configuration than yast is.
I dont hate suse, it just didnt fit our needs, and upgrades failed way too easily. these machines we need, they are critical. I could not have a machine fail after an upgrade. after several of these failed upgrades i said enough is enough and switched the suse machines out with gentoo machines.
so, that why I switched it out.
it needs some more polishing, but granted some of my own personal preferences were trampled by it and thats part of it, for sure.
I will try it again in the future, I keep my eyes on most ditros, except for micro$oft linux. (redhat)
but between debian and gentoo and upgrading, I think its a tie. both do it very well, and with very little pain.
suse left me hanging more than once with a trashed system. either it didnt boot or it was really b0rked.
I say, give it a shot. it works fine as long as you dont have like bleeding edge hardware.
(like some sata controllers, ect)
All the comments on how Netware was so great make me laff. Read "In Search of Stupidity: Over 20 Years of High-Tech Marketing Disasters"
by Merrill R. Chapman, to get a good feel for why Netware went by the wayside. Its wasn't the 'suits' that killed them, it was the engineers. They had a bunch of egotisical engineers who refused to build what the market wanted. Yes, it is/was a fast stable file server. But thats about it. You couldn't really run apps on it. And when the clients of the world are all running Windows, and Netware is expensive and difficult to use with Windows, there you go.
Having done some research on Novell about 6 months ago, there's some things that are not pointed out in the article that relate directly to cashflow.
It is true that their share of server sales has dwindled from a high of 80% down to low double digits. You can actually see that the impact of the introduction of WinNT. Over more recent years, the decline is actually due to Linux. What people frequently misunderstand is that this relates to NEW server sales.
What is not mentioned is the actual installed base. In this arena, Novell is huge. IIRC, a study that I saw put it at something like 3 to 5 million machines are running Netware.That's far more than all *nix combined, and was only matched by Windows.
Installed base == support/maintenance/upgrade revenue - ie services revenue. Novell has a cash cow there, that's for sure. Check out their 10Qs. And I should note that all their other 'businesses' - be it identity or ex Collabra or whatever are basically immaterial in comparison to their "software division formerly known as Netware".
But that revenue is still not sufficient. It was clear that if they didn't buy SUSE, they would lose their installed base to Linux for file/print services over time. And BTW, the Linux NEW server sales have been increasing rapidly. I always see that Linux is increasing at the expense of Windows and/or Unix, but never Novell. That is what surprised me most in my travels....
Obviously, migrating existing customers from Netware to Linux doesn't prevent leakage to either Windows or other Linux flavors, but now, they at least have a clear path for customers, and it will probably be the easiest path for their installed base.
Anways, that's my rant. And I wish Novell the best of luck. Successful linux companies will mean even better corporate acceptance.
So does Anonymous Coward have good karma?
Redhat is fine for servers (although I've had my share of problems with RHAS3, mainly NFS), but Novell offers something that RH has NEVER offered: out of the box Flash, Mpeg, Java, Real (yeah, I know - they're supposedly evil). These things are actually important for a desktop users. True, you could go out on your own and find all thiese things and install them yourself, but for a corporate "Give me a decent out-of-the-box desktop setup", Novell does nicely. And in my experience, Wireless as well as ACPI just work with MANY more laptops than RH.
For Novell, I think the biggest challenge is to keep revenue stable while customers transition from NetWare to Linux, without losing too many customers to Windows in the process.
NetWare is still pretty expensive on the server. A 50-user copy is about $150 a seat on CDW retail ($7,500), about $50 a seat under a licensing agreement ($2,500).
SuSE is $999 per server with no client licenses fees.
Figuring NetWare to be about 50% of Novell's one billion in revenue, that means Novell would stand to lose more that 25% of their total revenue assuming everyone switched to SuSE. Novell might make this up with SuSE/Ximian desktop revenue, but I see large amounts of revenue from Linux on the desktop as being a long time in the making.
The estimates for SuSE revenue for 2003 were for about $40 million in revenue. As near as I can tell Ximan never really made any money to speak of.
So, if I haven't bored anyone to death yet, Novell NetWare is a $500 million revenue stream, SuSE is a $40 million revenue stream. Novell needs to very carefully transition from NetWare to SuSE if they want to keep revenue even. They can also grow by taking customers from Microsoft or Red Hat. But, it appears to me that Novell will have to shrink about 25% in size in order to remain profitable in the short term. Red Hat, with a more mature Linux strategy, only made $100 million in the last four quarters.
None of this is a bad thing, and I wish Novell the best of luck. I used to work there, and I still have friends there. Just doing the math though it seems like they will need to get smaller before they get bigger again.
'I am become Shiva, destroyer of worlds'
Agreed.
1 755222). Here's another interesting link http://www.linuxworld.com.au/index.php/id;16020781 22;fp;16;fpid;0
Being Novell they have street cred. Netware gives them a wedge into big businesses which no other Linux distro has. And I think they know and understand the businesses' needs better than any of the others, and have the tools to complement Linux to cater to those companies (Zen, identy management and such).
I think Novell and SUSE make a good fit. SUSE now has a desktop platform which they can work WITH, and not AGAINST.
They're clearly aware of it, and their CC EAL4+ certification was part of their plan, a certification which I think only SUSE still holds (among the Linux distros).
What the article could have mentioned is that Novell is proving to their customer that they can do without Windows, migrating internally to Linux desktops (see Joe Barr's http://www.newsforge.com/article.pl?sid=05/03/23/
Not an urban legend. We (IT) lost track of a NetWare 3.x server used by a small (less than 100) department in a large (more than 20,000) entertainment company.
When it was time to upgrade the department's LAN from TokenRing to Ethernet, we couldn't locate this server. The clients had no idea where it was. They only knew that the all-important J: drive on their PC was there before they were.
It turns out that there used to be a janitor's closet behind shelves, and a large desk pushed up against the shelves, that no one in that department had ever remembered being opened. Certainly no one had the key for it. When it was finally opened, there was the 3.x, chugging away without a care in the world. I don't remember what the box was, but I doubt it held more than 16 megs of RAM.
By that time, the company had pretty gone the Microsoft route, and we were all weary of dealing with the christmas lights like nature of a WindowsNT based network. Needless to say, we were amused and more than a little impressed by this little-engine-that-could.
= 9J =