A Slightly Skeptical View of Major Commercial Unixes

News

Recommended Books

Recommended Links

 Reference HOW-TOs
Solaris Red Hat Suse AIX HP-UX
 

The biggest recent development in commercial Unixes space is that Oracle pledged to support Red Hat Enterprise Linux -- charging less than half of Red Hat's support prices ($99 for minimal level -- patches and RPMs only).  See:

As a news this was certainly on par with Google buying UTube for $1.6B hoax, which might be Sequoia capital trick of selling one of its properties to the other making 2 billions on 15 million investment in the UTube and helping Google to maintain its exorbitant share price a bit longer; or for a change with IBM hoax of buying ISS for over a billion. Still this is was an interesting development that reminds dot-com bubble mergers and acquisitions. Reminds us that the traditional Open Source (GPL-based Open Source), is a very problematic business model. It is often used as a smoke screen for the VCs to get software engineers to work for free, NOT EVEN Minimum Wage, but for free!  Then they take their hard work, pump $20M in marketing and either sell the resulting company to one of their other portfolio companies or take it public and dump the shares on the public. Meanwhile the software engineers that worked to develop that software for FREE, aka Slave Labor, get $0.00 for their hard work while the VCs and Investment Bankers make a killing. And of course then they get their buddies in Big Media hype the GPL-based Open Source id the best thing after sliced bread.

Oracle will also supply a RHEL clone. Oracle pledged Linux support from its own army of employees--including several Linux kernel programmers. Internally Linux is used by Oracle as a development platform along with Solaris. That makes the move somewhat similar to Sun buying Star Office: cutting internal costs. 

Anyway a good news is that the price of Linux support will be 50% lower and RHEL AS support model with its fat margins essentially died by quick painless death: customers now have a viable way to drive down the cost of maintaining RHEL and are likely to use this information in contract negotiations, even if they don't switch.

Eventually this move might divert major part of Red Hat's support subscription revenue stream into Oracle coffers.

Along with Red Hat, Novell is also under new pressure.  There is no longer much sense to use Suse as an enterprise distribution as its main selling point was lower support costs in comparison with Red Hat.
Notes:
  • Those pages are written by people for whom English is not a native language. Some amount of grammar and spelling errors should be expected.
  • This is a Spartan WHYFF (We Help You For Free) site. It cannot replace the best teachers and the best books.
  • The site contain some obsolete pages as it develops like a living tree... Some links on older pages are broken. Please try to use Google, Open directory, etc. to find a replacement link (see HOWTO search the WEB for details). We would appreciate if you can mail us a correct link.

Search Amazon by keywords:

 Google   
Open directory
Research Index
 

Old News ;-)

[May 18th 2007]  freshmeat.net Project details for iogen by Marco Peereboom

About: iogen is an I/O generator. It forks child processes that each run a mix of reads and writes. The idea is to generate heavily fragmented files to make the hardware suffer as much as possible. This tool has been used to test filesystems, drivers, firmware, and hardware devices. It is by no means meant as a performance measuring tool since it tries to recreate the worst case scenario I/O.

Changes: This release adds a timeout parameter for I/O runs.

Red Hat What me worry Between the Lines ZDNet.com

Dan Farber & Larry Dignan
December 21st, 2006

Red Hat: What me worry?

When Red Hat holds court with financial analysts later today to discuss the company's fiscal third quarter results the conversation is likely to go like this:
Analyst: What is the impact on Oracle's Unbreakable Linux on your business? How can you compete?

Red Hat exec: We're not seeing any direct threat. Billings are looking up.

Analyst: What about this Microsoft-Novell partnership?

Red Hat exec: Can't we talk about our quarter just a little here?

And then you'll get a lot of questions about Red Hat's forecast for future billings so analysts can surmise the answers on their own. Red Hat, which provides Linux and open source software, is expected to report earnings of 12 cents a share on sales of $104 million.

The competition is circling around Red Hat, but it's too early to see the effects. Are folks going to jump from Red Hat? Possibly, but it won't be this quarter. Or the next.

How do I know? Let's evaluate what those aforementioned Red Hat killers are saying lately. Oracle said it had 9,000 downloads of Unbreakable Linux in the first 30 days after announcing it. Big question is what happened beyond that 30 days and were the downloads front-end loaded indicating waning interest.

As for the Microsoft-Novell deal, the two parties announced that three financial services firms are getting SUSE Linux Enterprise Subscription certificates from Microsoft.

What's all of this mean for Red Hat? Probably a decent quarter after a lot of worrying. What's it mean for technology buyers? Some serious leverage as Red Hat subscriptions expire in a few months.

UBS analyst Heather Bellini said it will take about six months to see any dent in Red Hat's business.

"Our conversations with Linux channel partners indicate that Red Hat's business in the third quarter was largely unaffected and any impact will take at least six months to play out. While the resellers were equally split on whether Red Hat will have to lower prices, we believe pricing pressures are inevitable as customers will at the very least use Oracle's pricing to negotiate deeper discounts."

WR Hambrecht analyst Robert Stimson said in a research note that Red Hat's products are "sticky and deeply embedded within its enterprise customers."

"We are expecting a roughly in-line quarter to both our and Street estimates, as we believe recent concerns over competitive pressures from Oracle and Novell/Microsoft have been overblown. We believe investors will be most closely focused on billings growth as the most meaningful metric to determine any negative effect from competition, as well as management's commentary on the JBoss integration process."

In fact, the integration of open source software maker JBoss is the more immediate worry about Red Hat. Rumors have swirled for weeks about Marc Fleury leaving, but thus far no formal announcement.

The skinny from Stimson:

"Comments from JBoss head Marc Fleury in late November regarding a perceived lack of investment from Red Hat raised investors' concerns about the integration of the JBoss business into the Red Hat stack. Shortly after, Fleury took paternity leave, which some investors read into as a sign of discord. Although we believe these concerns are overblown, we will be looking closely at commentary from management regarding the progress of the JBoss integration."

[Dec 21, 2006] Linux Today - OSNews Has the Desktop Linux Bubble Burst

Linux Today is not responsible for the content of the message below.
Don Ferguson - Subject: Excellent Points ( Dec 21, 2006, 17:07:53 )
I am no fan of glossier, fancier and more processor and memory intensive user interface graphics. But a lot of people are. More importantly, there is real movement on the Mac and Windows front from a user interface perspective and these changes will define computing experiences and desires.
I do not think Windows Vista is a Linux killer. But, Leopard is. Apple has proven you can do *nix and make it a visually pleasing, easy to use, and productive environment for people. Macs run Linux programs, Mac programs, and Windows in VMs or as a dual boot option. Apple has created everything that Linux ever wanted to be but couldn't achieve because it was unable to attract enough people to the KDE and/or Gnome.

I almost wonder if what we are seeing with KDE and Gnome is symptomatic of "open source". The movement relies on coolness to bring in new, young coders, and its anti-establishment exture to bring in rebels.

Linux, KDE or Gnome are getting less cool by the day. And Linux and several open source projects are so tied into corporations that the work of rebels is simply being used to increase the fortunes of billionare wannabees.

Slashdot Oracle and Red Hat begin battle for the Enterprise

  • Is this the end of the OSS "Sell the Support" mode

    by demallien2 (991621) on Monday October 30, @08:54AM (#16641085)
     
    It has always seemed relatively obvious to me that most OSS software companies are vulnerable to this type of attack mounted by a large proprietary software vendor. Take the software (which, at the end of the day is where the real value is), and offer support, but without undertaking any of the major development tasks (only do bugfixes). The OSS competitor has two choices: continue to do R&D work on the product, to keep it advancing, and accepting that they can't sell support as cheaply as the "bug-fix only" proprietary vendor, or stop doing R&D themselves, so that they can be cost-competitive. Of course the disadvantage of this approach is that the product quickly falls behind proprietary offerings....

    This is not going to be an easy battle for Redhat. I suspect they are going to have to find a new business model if they are to survive.

  • A business owner disagrees

    by NineNine (235196) on Monday October 30, @10:36AM (#16642345)
    (http://ninenine.com/)

    I'm not so sure the real value is in the software. People and, especially, companies seem to be willing to pay more for support contracts than for software. They'll even take inferior software over superior software if they can get a support contract that way.
     
    I own a small-ish business. In no way, shape, or form, is support more important to me than quality software. If I have to make support calls, that's lost time and money. The second software malfunctions, is the second you start losing money. No question about it. I will pay multiples more for a product that requires little to no support, than I will for a product that has good support.

    The only reason a product needs support is because it's not good enough to be used without support. So by definition, a better product requires less support than a product (that does the same thing) that requires support. There's no value in support. Support is purely a cost, and an avoidable one at that.

    In the case of somebody like Red Hat, there's simply no way I'd ever use the product (at least for our desktops... our server stuff is outsourced). I don't care if I can get a literate, English speaking person on the phone instantly 24/7 via a toll-free number. I don't care if the company will teleport a support person to my company within one minute of needing help. That's not nearly as good as using a product that doesn't require support.

[Oct 28, 2006] Perspective: Sending the penguins out with a hot foot

This rates as a stroke of cruel genius--right up there with Bill Gates' decision to gut his Borland nemesis Philippe Kahn in the early 1990s. Old-timers may recall that Borland once was a high-flier in the software business. But when Microsoft slashed prices on its Excel spreadsheet and Access database programs, Borland stumbled. The company failed to find a way to compete against a bigger, better financed rival that could afford to pursue a beggar-thy-neighbor strategy.

Hand it to Ellison for taking a page out of his arch-rival's playbook. Oracle's offer of free support for Red Hat Linux was designed to inflict maximum pain on Red Hat. So it did. One day after the announcement, Red Hat shares lost 24 percent of their value. After watching his stock take a tumble, Red Hat's CEO Matthew Szulik is in a bind. He has just absorbed the equivalent of a cyber-kick in the groin from a bigger, badder bully.

"This is capitalism, we are competing," Ellison later said during the question-and-answer session following his announcement. "We are trying to offer a better product at a lower price."

It's also a veiled threat to any open-source software vendor within earshot that Oracle's declaring a support price-war. Outside of an IBM, I don't know of any open-source supplier with the financial wherewithal to absorb that kind of profit margin punishment.

[Oct 28, 2006] Will Red Hat Survive#16621552

  • Fedora will never be a production OS

    (Score:1, Informative)
    by Anonymous Coward on Saturday October 28, @08:09AM (#16621534)
    Fedora on the other hand is free.
    There's a good reason Fedora is free. It's not a production OS and it never will be, for that would conflick with RedHat's ability to sell its "enterprise" products. You can use Fedora if you want to debug problems for RH for free.

    As Bruce Perens said it a while ago:
    Fedora project is obviously intended to look like Debian. But unlike Debian, Fedora is an extremely unequal partnership. "Fedora" is where the community developers are supposed to build Red Hat's product, while the certifications and vendor endorsements are held back for the high-priced "Red Hat Enterprise Linux" brand. This is especially obvious in recent certification announcements: the Common Criteria certification will go to "Red Hat Enterprise Linux", not "Fedora". And of course the entire steering board of the Fedora project are Red Hat employees. Red Hat recently announced a second draft of the leadership structure for Fedora, in which they have eliminated voting, expressing the need to keep control in the hands of Red Hat's management.

    But the most ludicrous aspect of the Fedora project is that with Fedora, Red Hat seeks to achieve what Debian did long ago. Because they can't (and shouldn't) control Debian, they decided to re-invent the wheel. It would take them years to achieve a fraction of what Debian already has.
    If you need a stable, easy-to-administer, well-established production OS, I would suggest Debian [debian.org].

  • Re:Prices--Red Hat way more expensive

    (Score:3, Informative)
    by lukas84 (912874) on Saturday October 28, @10:50AM (#16622560)
    (http://projectdream.org/)
     
    I work in sub-200 people environments in switzerland. So my perpspective might differ a lot from yours, but i've found microsoft to offer good value for their money.

    1 CHF (Swiss Franc) = 0.80 US$

    If you're a sub-15 people company, and only have one machine, Microsoft Small Business Server is a good bet. You can get the Standard Edition for 800 CHF, and the Premium Edition (includes SQL Server and ISA Server (no idea why you would want that)) for 1300 CHF. Each includes 5 User CALs, So for 15 people you need two more five packs or about 1000 each. This price doesn't include support, though. Also, i'm not much of a fan of SBS because of several restrictions (only a single domain controller, Exchange and DC on the same machine), but these are the standard practice in such small companies.

    Microsoft offers their SBS server for up to 75 users. I don't think thats a good idea.

    Companies in this size usually don't have any IT staff, so self help is important. With windows, the people at least feel that they can try to fix problem themselves (which they usually can't). With Linux, this isn't the case. (Just because windows server offers a GUI for 80% of it's functions doesn't mean that it's simple).

    Windows is mostly the only choice if you are cooperating with other companies. Some might offer their shipping calculation program only for windows, some specialized ERP software might only be available or windows, etc. pp.

    I've found windows to offer the best SMB desktop management, everything from redhat etc. seems to be geared at big companies with a standard desktop images. Group Policies are a fucking cool thing.

  • Re:Prices--Red Hat way more expensive

    (Score:2)
    by lukas84 (912874) on Saturday October 28, @02:15PM (#16624188)
    (http://projectdream.org/)
    The answer is, of course, it depends.

    If you are a information technology company, it might be possible and might even have lots of advantages to use linux exclusively.

    However, if you are not an IT company, i don't see linux to stand a chance. There are some technical reasons here, but also social reasons:

    The technical reasons:

    * Software, which the company needs might not be available under linux. Using VMs or WINE might solve the problem in the short term, but what if the a new version doesn't work anymore?
    * Especially for low end hardware, there isn't much support for linux. But if we're honest here, low end hardware causes lot's of trouble under windows.
    * Integration with linux is very difficult. All kinds of mobile phones, pdas, mp3 players, etc. can be hooked up to windows with the help of a CD. With linux, it isn't as easy. While some devices might be supported, some of them won't be. Unfortunately, this also includes high end hardware (like Windows Mobile PDAs).
    * Groupware with Linux seems to be a problem still. I lack experience in this area, but last time i checked solutions here weren't as well integrated (for example lacking support for Direct Push, Blackberry, etc.)

    The social reasons:

    * There usually are no people with linux knowledge ("Power Users"). This means for every so little problem, external support is required, adding a lot to the cost of linux. A technicians hour is usually the same as a single CAL. This problem will solve itself over time, though.
    * People are afraid of Linux. It is new and unknown. People like to have something to blame. If they can't get their document out because they have to use Linux to write it, they will blame Linux. Irregardless of facts.

    As it is now, linux can't be adopted by small companys. Larger companys have much more resources available, and might save money when deploying linux.

  • Re:Prices

    (Score:2)
    by msobkow (48369) on Saturday October 28, @09:16AM (#16621876)
    (Last Journal: Saturday October 28, @09:32AM)
    Ok, so RedHat provides buzzwords and certs so managers and business owners can comfort themselves with the warm fuzzies that their techie has a cert. Is the cert really that much better than Microsoft's various techie certs? Has a cert ever actually demonstrated someone has the skills and training needed to do the job right, or does it just prove someone had the time and money for a course?

    Clearly certs aren't enough to maintain a company of RedHat's size, or there would be major competition from companies like Learning Tree by now.

    Security audits of code, patches, updates, enhancements -- those are what a vendor is supposed to be focused on delivering. The problem is that with OSS, the benefits of those corrections go to everyone, but the expense only goes to the company that developed the fix or enhancement (and their customers.)

    SuSE/Novell put in a lot of time, money, and effort. RedHat put in a lot of time, money, and effort. So has Mandrake, whatever their latest name is.

    What has Oracle put into Linux?

    Show me the pieces of system code that have been enhanced and updated by Oracle. I want to know which security issues and performance tweaks they've implemented. That trail of invested effort will show how good their team is at providing service and support.

    If they haven't invested the time, budget, and effort before someone paid them to provide support, why in the world would I trust a database vendor to maintain my operating system?!?!?!

 

  • Re:They Don't Get It

    (Score:1)
    by Lord Kano (13027) on Saturday October 28, @04:44PM (#16624742)
    (http://www.angelfire...epublican/index.blog | Last Journal: Thursday July 27, @01:00AM)
     
    Despite what a few whiners around here say, Redhat supports the community and takes on people who attack it. They have built tremendous good will.

    Red Hat burned nearly as much good will as they currently have remaining.

    Red Hat's high price of entry makes them less important to the rank and file in the OSS world than they used to me.

    I was a Red Hat user from 4.2-7.3, and since I am not interested in sparking oh yet another distro jihad, I'm not going to go into what my current distro of choice is. That whole "Let the community be our Beta testers while we sell the mature product to the important people in "Enterprise" " mindset has lost Red Hat a lot of the good will that they built up in the time before.

    LK

    • Probably not

      (Score:3, Informative)
      by cpbrown (794387) on Saturday October 28, @06:20AM (#16620956)
      (http://cpbrown.deviantart.com/)
       
      I don't think that it will. It is one of the original heavyweights but in the face of newer and more specialised distros it no longer occupies a suitably small niche in order to ensure its long term continuation.

      In my opinion, most serious developers will keep to a lighter distro, and most newbies will keep to a nice flowery distro such as Ubuntu, which prides itself on ease of use. Red Hat is no longer necessary. Competition will inevitably drive it away in the ever dynamic food-web of free software.

  • Why didn't Oracle simply buy one of the groups...

    (Score:1)
    by g2racer (258096) on Saturday October 28, @08:12AM (#16621316)
    (http://www.tedshouse.com/)
    That already respin from the RHEL srpms? Seems like they now have to build a community for "Unbreakable". Shouldn't be difficult given who they are, but why start from scratch?

    Interesting strategy though. Wonder if this is payback for JBoss?

    Definitely don't think that RH will go into bankruptcy any time soon. They've got a rich product stack and plenty of customers. The only thing their really lacking is an enterprise database platform. May RH should consider buying Sybase or the Informix stack from IBM so they could go head to head with Oracle ;)

  • Never mind Red Hat...

    (Score:4, Insightful)
    by melonman (608440) on Saturday October 28, @08:15AM (#16621326)
    (Last Journal: Saturday April 12, @08:08AM)
     
    Surely the bigger loser here is Novell. Oracle is competing with Red Hat for support of RHEL, but then Red Hat never had the monopoly in that market anyway: plenty of people used Red Hat without paying for AS-quality support. What this move does do is make the Red Hat flavour of Linux even more clearly the mainstream enterprise distribution. How well Red Hat will cope with competition from Oracle in offering support for that product remains to be seen, but I'd have thought that selling a non-Red Hat flavour of Linux to an IT department suddenly got a whole lot harder.

  • Re:Will Oracle Survive?

    (Score:2)
    by Decaff (42676) on Saturday October 28, @11:11AM (#16622240)
     
    Now the FLOSS database solutions are reliable and sophisticated enough that exansive solutions such as Oracle are only required by niche applications. The only thing really keeping them afloat is intertia: there are a number of talented Oracle DBA's who would like to continue using what they know; and there is a stable of important applications which will continue to certify Oracle for some time. It will take time for existing application vendors to migrate to commodity backends, but it is very much in their best interest to do so.

    I think this shows a fundamental misunderstanding of the database market. Oracle is not niche - it is mainstream. It is one of the databases that a company uses when it must have very high uptimes and guaranteed performance. Companies that do on-line commerce are not like blog sites or forums, where the occasional delay or timeout is merely an inconvenience. For even smaller companies, the cost of Oracle is trivial compared to company turnover or staff salaries.

    I really love PostgreSQL and use it for most of my development, but there is no way I would use it to host a high-performance site running financial transactions. PostgreSQL may well be good enough, but the key word is 'may'. I can be sure that Oracle is.

[Oct 26, 2005]  Oracle as new Robin Hood

#16594350)
That Unfakeable page is act of desperation...
Red Hat spreading FUD about another open source product, how noble! And let's not forget how they sent that cease-and-desist letter for CentOS for stating they're based on RHEL...

Let's see what they have to say:

Q: Does Oracle's announcement include support for the Red Hat Application Stack, JBoss, Hibernate, Red Hat GFS, Red Hat Cluster Suite, Red Hat Directory Server, or Red Hat Certificate System?
A: No. Oracle does not support any of these leading open source products.
-
Uhm, that doesn't matter.

The point is this: for any any Unbreakable Linux bug that is submitted to Oracle and can be duplicated on "golden" RHEL 4 system in Oracle's office (for which Oracle has valid support contract), Oracle can submit it to Red Hat Support as Red Hat bug and require quick fix. Then, as RH fixes it, they can fix it in their own Unbreakable Linux.

Q: Oracle says their Linux support includes the same hardware compatibility and certifications as Red Hat Enterprise Linux. Is this true?
A: No. Oracle has stated they will make changes to the code independently of Red Hat. These changes will not be tested during Red Hat's hardware testing and certification process, and may cause unexpected behavior. Hence Red Hat hardware certifications are invalidated.
--
Well, yes. They don't say that Red Hat h/w certs will be considered valid (actually, they don't care, to be exact) - as long as you've got one, they'll support your RH cert on their Unbreakable Linux. The same goes for ISVs. And Oracle isn't that stupid to screw things up so that they don't work.

Q: Oracle says they will provide the same updates as Red Hat Enterprise Linux. Can they do this?
A: There are multiple requirements to building binary compatible software. One piece is the source code; another is the build and test environment. While Oracle may be able to take the source code at some point after a Red Hat update release, obviously their build and test environment will inherently be different than that for Red Hat Enterprise Linux. For similar reasons, there is no guarantee that the source code for the Red Hat Enterprise Linux update will work correctly when integrated into Oracle's modified Linux code base.
--
Hah, this one is hilarious! So this is Red Hat's secret sauce - the unreproducible build environment. Are they trying to say that their build environment is different from what's available to everyone else (which wouldn't be too good for compatibility which they emphasize all the time)?

Q: Does Red Hat allow you to tailor your support level to your workload?
A: Yes. Many customers match their Red Hat Enterprise Linux subscription level to their application SLA requirements. For example, customers may choose a Basic subscription for non-mission critical file and print servers, while selecting Premium subscriptions for database servers. Oracle does not allow this flexibility - their support policy reads: "If acquiring Enterprise Linux Premier Support, all of your Oracle supported systems must be supported with Enterprise Linux Premier Support."
---
Nice try. On the other hand Oracle's Linux is free and updates only are $99/year. Match that, Red Hat! Basic support for RHEL Workstation is $279.
As for Oracle DB servers - yes, you'd probably want to have premium Linux support for those.
BTW, did RH mention that their support agreement requires that support must be purchased for all copies of OS used by the customer?
Self-tuning SLAs can also be achieved by using CentOS (community and basic support), RHEL and UL.

Q: Can Oracle produce timely security updates to Red Hat Enterprise Linux as they stated?
A: No. There will be a delay between the time a Red Hat Enterprise Linux update is issued, and the time the source code makes its way to Oracle. And there is no guarantee that the source code for the Red Hat Enterprise Linux update will work correctly when integrated into Oracle's Linux code base; this integration and test may take additional time. In the case where the update corrects critical security flaws, Oracle customers may be exposed to additional risk.
--
Yes, if the bug is submitted to Red Hat, there might be a delay of 1-2 hours.
If it's submitted to Oracle or to CentOS, RH and Oracle, there's no reason why Oracle couldn't issue their own fix before RH and, if change doesn't require reboot, re-issue RH's update after they get it from Red Hat. It's great to know that Linux requires timely security updates because it seems prone to frequent critical security problems, though.

Q: Will Oracle's Linux customers have the same degree of influence over Oracle's Linux as Red Hat's customers do with Red Hat Enterprise Linux?
A: The support we provide for Red Hat Enterprise Linux starts when Red Hat and its customers collaborate in the design of new versions. This collaboration extends through the development, test, and production deployment of Red Hat Enterprise Linux. Vendors of a derivative distribution are simply not positioned to provide their customers the same collaboration opportunity.
-
Oh yes, Red Hat is well-known for their excellent collaboration with ISVs and IHVs... And they're very easy to work with...

Q: Hardware vendors such as Dell, HP, and IBM provide support for Red Hat Enterprise Linux. How is Oracle's support offering different?
A: Red Hat's hardware partners provide front line support to customers, backed by Red Hat. Red Hat has a close contractual relationship with these partners, which requires training, well defined escalation paths, Red Hat back-line support, and cooperative customer issue management. Our joint customers enjoy the same degree of collaborative participation as any Red Hat customer.
--
According to HP's stats (you can google news.com for that article) in 2005 about 4,000 Linux support issues escalated to the HQs, less than 100 had to be escalated to Linux vendors.
Besides, all major OEMs endorsed Oracle's Unbreakable Linux, see today's news.com article on UL.
Maybe they don't give a damn but hey - why not get yourself in a position which helps you get a better price for RHEL :-)

In any case, each dollar invested in UL makes RH cheaper by more than one dollar, so this is a nice move by Larry. So far, so good - (http://finance.google.com/finance?q=RHAT - not a pretty sight).

And isn't it great that someone can take the Fedora -> RHEL model, where RH profits from work done by others (open source community, in RH's case) and make it work for the enterprise customer (RHEL -> UL) while profiting only from the rich. Unbreakable Linux = Robin Hood Linux :-)

[Sep 14, 2005] Dr. Dobb's Red Hat Releases Enterprise Linux 5 Beta September 13

... The Red Hat Enterprise Linux 5 Beta 1 release contains virtualization on the i386 and x86_64 architectures as well as a technology preview for IA64.

... ... ...

Aside from Xen, Red Hat Enterprise Linux 5 Beta 1 features AutoFS and iSCSI network storage support, smart card integration, SELinux security, clustering and a cluster file system, Infiniband and RDMA support, and Kexec and Kdump, which replace the current Diskdump and Netdump. Beta 1 also incorporates improvements to the installation process, analysis and development tools SystemTap and Frysk, a new driver model and enablers for stateless Linux.

IBM Redbooks Linux Client Migration Cookbook A Practical Planning and Implementation Guide for Migrating to Desktop Linux

The goal of this IBM Redbook is to provide a technical planning reference for IT organizations large or small that are now considering a migration to Linux-based personal computers. For Linux, there is a tremendous amount of “how to” information available online that addresses specific and very technical operating system configuration issues, platform-specific installation methods, user interface customizations, etc. This book includes some technical “how to” as well, but the overall focus of the content in this book is to walk the reader through some of the important considerations and planning issues you could encounter during a migration project. Within the context of a pre-existing Microsoft Windows-based environment, we attempt to present a more holistic, end-to-end view of the technical challenges and methods necessary to complete a successful migration to Linux-based clients.

Jun 24th, 2004 Open Source Blog: Open Sourcery by Blane Warrene

I recently spent some time speaking with a popular Yankee Group analyst who covers the enterprise sector in the US, focusing in on open source and where the movement may go in the next few years.

Just to be clear, I differentiate, as most industry watchers do, between Linux and open source. While Linux is open source, the primary Linux distributors have caught on to how they need to position themselves for success and are starting to run their businesses just as any proprietary software company does.

Red Hat and SUSE make prime examples, realizing the path to long term success and revenue streams resided in proving themselves enterprise worthy to larger businesses and institutions, have shifted business models or been acquired by organizations with roots in the enterprise.

Her views, while not always popular in the open source community. are right on point if open source seeks widespread adoption and a permanent seat at the table for longer term financial success.

There are a few obstacles open source proponents need to accept and move forward on:

  1. It will be more costly for a company to migrate away from Windows to Linux, even in light of slightly reduced ongoing maintenance and improved security and uptime. While I have not always agreed that the costs are higher, having migrated corporate systems to Linux in the past, their research showed it to be true in many cases -- especially when migrating beyond standard web hosting and email systems. The costs are higher when factoring in re-certifying drivers, application integrity and training.
  2. To truly become entrenched as a viable financially-rewarding option (meaning open source companies make money and create jobs), a shift toward commercial software models is necessary. This does not mean forgoing open source, however, what it does mean is developing a structure for development, distribution, patching and support that passes muster with corporate IT managers who could be investing substantial amounts of money in open source.

What it boils down to is that while open source has definitely revolutionized software, and it is found internationally in companies large and small, businesses still pick software because it provides a solution not just because it is open source.

The fact that it is cheaper or free simply means the user will save money, but this does not win the favor of those buyers who could be injecting millions into open source projects rather than proprietary software makers.

I would use Firebird as a model. In an interview with Helen Borrie, forthcoming in my July column on SitePoint, she noted that since many Fortune 500 companies are using an open source database like Firebird speaks volumes to the maturing of their project and open source at large.

The reason as I see it, is due to the treatment of Firebird like an enterprise scale proprietary software project. They have a well managed developer community and active support lists, commercial offerings for support through partnerships with several companies, and commercial development projects for corporate clients.

If more open source projects looked at Borrie's team model and discipline in development and support, we just might see more penetration that attracts longer and more profitable contracts and work for those like us in the SitePoint community.

(Post a comment)

Comments

It will be more costly for a company to migrate away from Windows to Linux, even in light of slightly reduced ongoing maintenance and improved security and uptime. You mean relative to staying with Windows? Does this include recurring costs of Windows licensing / upgrades?

The costs are higher when factoring in re-certifying drivers, application integrity and training.

On the drivers front, that assumes (if we're saying Linux cf. Windows) that systems need upgrades as frequently. There's generally less need to keep upgrading Linux, when used as a server.

Re application integrity, think thats very hard to research accurately - kind of a wooly comment that needs qualification.

On the training side, it's an interesting area where it's kind of like comparing Apples with Pears.

Windows generally hides administrators from much of what's really happening, so it's probably easier to train someone to the point where they're feeling confident but given serious problems, who do you turn to?

*Nix effectively exposes administrators to everything so more time is required to reach the point where sysadmins are confident. Once they reach that point though, they're typically capable of handling anything. The result is stable systems. I'd also argue that a single *Nix sysadmin is capable of maintaining a greater number of systems (scripts / automation etc.) although no figures to back that.

Firebird is an interesting example. The flip side of Firebirds way of doing things seems to be the Open Source "community" is largely unaware of it (compared to, say, MySQL).

  Posted by: HarryF from phppatterns.com Jun 24th, 2004 @ 8:03 AM MDT

Comment

Yes - on costs - Linux was actually found to be more expensive in numerous cases compared to staying with Windows. This is unfortunate as I am a proponent of finding migration paths from Windows to Linux for stability and administration automation. However, the research did show the total cost of ownership eventually balances out, it simply is much more expensive at the outset than staying on a Windows upgrade path.

This survey (partially on site with staff and others via questionnaire) - 1000 companies with 5000 or more employees - found that they did have to certify drivers at the initial migration, certify all new disk images, provide training or certification to adhere to corporate policy, buy indemnification insurance, perform migrations, test, establish support contracts and finally, pay about a 15 percent premium when bringing in certified L:inux staff.

The benefit if the company decided to take the financial hit: over an extended period they experienced the benefits of Linux - uptime, experienced admins and flexibility of the platform.

Application integrity was ambiguous in the study - however - managers cited it constantly when trying to retire commercial Unix and move apps to Linux, needing certification that an entire applications runs exactly as before.

Perhaps it is time for the open source community to begin establishing central organizational points that act as clearinghouses - like Open Source Development labs does for Linux - to certify open source applications on a major scale.

  Posted by: bwarrene from practicalapplications.net Jun 24th, 2004 @ 1:12 PM MDT

Comment

I beg to differ on Harry's view about Firebird. Firebird is not as popular as MySQL because 1) it's a newer project (project, not software) and 2) MySQL support comes built into PHP; no need for additional software. Firebird requires either recompilation or loading this DLL into the extension space.

  Posted by: andrecruz Jun 24th, 2004 @ 9:37 PM MDT

Comment

It was nice to read about your chat with L... DiD... (why are we keeping her name secret?).

Second, I don't understand your distinction between Linux and Open Source. Maybe I'm slow or something, but what it seems to boil down to is:

"Open Source = unprofessional Proprietary = professional (unstated) Linux = open source, but starting to become professional despite itself by acting like proprietary."

Well I'll grant you there are a lot of unprofesssional Free Software projects out there; but the same is true of proprietary. Bad proprietary programs are slightly less likely to see the light of day, but there's still a bevy of them out there.

Now, on the assertion that Linux companies are succeeding by acting like proprietary companies: there's truth and non-truth to it. On the one hand, Red Hat and SuSE have no doubt learned a lot about management, marketing, and good business practices from established companies. On the other hand, an effective open source player does not act the same as an effective proprietary player: there are all kinds of issues with dealing with the developer community that are not an issue in the proprietary world: they bring plusses and minuses, but have to be dealt with rather than ignored.

And I will note that Red Hat, the most successful Linux distributor, is a pure-play Open Source vendor: they do not ship proprietary code. In fact, they devote a lot of developer time to a community distribution that they make no direct money on (but do get free testing from). Likewise, one of the first things Novell did after its so-far successful acquisition of SuSE was to GPL SuSE's proprietary installer. This suggests that while good management is indispensible in anythin, Open Source ventures should not be running off and trying to ape proprietary vendors blindly.

Finally, there's a big difference between the way mass-market shrinkwrapped proprietary software and the way big-iron stuff is. With big-iron stuff you often have consultants in the field, lots of direct customer feedback, maybe even code sharing under NDA with the client: in short, it works a lot like an Open Source project. And that's where Open Source has shined: *nix boxes, web servers, network infrastructure, compilers, developer tools, and increasingly RDMSes. With mass shrinkwrap you have to do much more seeking out of customer needs on your own and also be prepared to tell customers to shove it and wait for the next release. On stuff like this (desktop guis and apps) Open Source has been less successful.

At least one high-profile OSS desktop project (Mozilla) was a legendary quagmire for a long time and is only beginning to claw its way back. Many of the mistakes came from not being open to community input ("dammit, we don't need a whole platform, just a good browser") as any good project of any kind should be. Thing is, no one has a clear idea of how to be usefully open to community input on a mass-market OSS project yet: the twin dangers of adding every requested feature or my-way-or-the-highway-ism have been so far hard to avoid.

Personally, I think the question of the Open Source desktop is given too much importance. Windows server shipments still account for 60% of the market, so it's not like that area is all sewn up. A company that wants to avoid vendor lock-in would do best to migrate its server infrastructure first - that's gonna be least painful and probably highest long-term benefit. Then maybe desktop apps, the maybe desktop operating system.

On MySQL vs. Firebird: yes, MySQL is more widespread, but they're used for entirely different things.

  Posted by: jmcginty Jun 25th, 2004 @ 12:34 PM MDT

Comment

I'm a bit confused to why you want to differentiate between Linux (eg. Red Hat) and Open Source.

Red Hat releases source packages and contributes largely to Open Source projects, both in resources as in code. Improvements by Red Hat are included in SuSE and vice versa. Everybody wins.

This ensures that Red Hat will have to be the best on its own merits. Competition will always be lurking around the corner to take over. Despite that, Red Hat is doing a good job.

You cannot compare this to proprietary vendors were your money goes into the big company bucket being used for the next version that you have to pay for again.

If I can choose I'd rather pay for services, if it guarantees that the money is used for Open Source development. If my Open Source vendor goes belly-up, its work is still available for anyone to use.

Paying for Open Source just guarantees you that you have freedom and are never tight to any vendor. Red Hat is just one example to show that the money is used for the good of the public.

And if you don't have deep pockets, there's still Fedora, CentOS, TaoLinux or Whitebox. Plenty of competition in the same vendor segment. Hard to beat IMO.

  Posted by: Dag Wieers from dag.wieers.com Jun 26th, 2004 @ 3:57 AM MDT

Comment

One thing I notice that is never mentioned when talking about Windows vs. Linux TCO is virus & worm costs. Both the cost of AV s/w and clean-up after an infection sneaks into the corporate LAN. That *huge* expense will never be borne by a Linux shop.

Posted by: Ron Johnson Jun 26th, 2004 @ 7:56 AM MDT

HP Throws Weight Behind MySQL, JBoss

HP Throws Weight Behind MySQL, JBoss
By Clint Boulton

HP (Quote, Chart) stepped up its commitment to open source software Monday by pledging to offer and support the MySQL database server and JBoss application server software in its servers.

The Palo Alto, Calif. systems vendor said it has inked agreements with those open source purveyors to certify and support MySQL and JBoss software on its servers.

Jeffrey Wade, manager of Linux Marketing Communications at HP, said the certifications factor in the company's Linux reference architecture is a software stack that covers everything from the hardware to the operating system, drivers and management agents.

Deployed on HP ProLiant servers, the open source Linux Reference Architectures are based on software from MySQL, JBoss, Apache, and OpenLDAP. The company's commercial Linux Reference Architectures are based on product from Oracle, BEA and SAP.

Both MySQL and JBoss will join the HP Partner Program and receive joint testing and engineering support on HP's hardware systems.

Wade told internetnews.com the added layer of MySQL and JBoss support addresses one of the largest concerns customers have today in opting to pick open source technology over mainstay proprietary products such as Microsoft (Quote, Chart)Windows, Sun Microsystems' (Quote, Chart) Solaris or UNIX.

"We can provide support for that entire solution stack and we're also now giving our customers flexibility in choice and the types of solutions they want to deploy whether that's a commercial or open source application," Wade said.

Bob Bickel, vice president of strategy and corporate development at JBoss, said commercial use remains somewhat constrained because a CIO doesn't know whom they can turn to for support.

"They don't know who they can turn to for indemnification," Bickel told internetnews.com. "Yeah, it works great and it's cheap but what happens in the middle of their big selling season if something goes down. Who do they turn to and get it from. What HP's doing is taking an all encompassing view of this with certification and testing."

Testing keeps customers from guessing what version of a Java virtual machine, operating system, MySQL or JBoss product can all work together in a guaranteed way, Bickel explained.

MySQL Vice President of Marketing Zack Urlocker said companies such as Sabre are using an open source stack for business applications. Partnering with HP, then, provides great validation for MySQL and JBoss software.

"A couple of years ago the big knock on open source was that it might be good on the periphery or Web applications, but was not quite ready for business critical applications," Urlocker told internetnews.com. "Now, the No. 1 issues have been support. People who have had a lot of success with Linux are now looking at how to use a whole open source stack."

The deal is truly symbiotic. While MySQL and JBoss get backing from a technology driver such as HP, HP gets the added credibility of being cozy with open source, a label many enterprises and HP rivals, such as IBM (Quote, Chart) and Dell (Quote, Chart), are working toward.

Linux sales are trending tall regardless; according to recent hardware server and database software studies from high-tech research outfit Gartner.

Despite legal threats from SCO Group and competition from Microsoft, Gartner said Linux continued to be the growth powerhouse in the operating systems server market, with a revenue increase of 57.3 percent in the first quarter of 2004.

Gartner also found that Linux siphoned market share from UNIX in the relational database management system (RDBMS) market, a niche that grew 158 percent from $116 million in new license revenue in 2002 to nearly $300 million in 2003.

Copyright © 1996-2008 by Dr. Nikolai Bezroukov. www.softpanorama.org was created as a service to the UN Sustainable Development Networking Programme (SDNP) in the author free time. Submit comments This document is an industrial compilation designed and created exclusively for educational use and is placed under the copyright of the Open Content License(OPL). Original materials copyright belong to respective owners. Quotes are made for educational purposes only in compliance with the fair use doctrine.

Standard disclaimer: The statements, views and opinions presented on this web page are those of the author and are not endorsed by, nor do they necessarily reflect, the opinions of the author present and former employers, SDNP or any other organization the author may be associated with. We do not warrant the correctness of the information provided or its fitness for any purpose.

Created May 16, 1996; Last modified: June 05, 2008