Red Hat CEO Matthew Szulik Responds

The "Call for Questions" post for Mr. Szulik generated over 650 questions and comments, 32 of which were moderated +5 at the time we selected questions to send him. And unlike many CEO responses, Szulik's answers to the selected questions are his own, not PR-generated. (One clue is that they are not in perfect English, as interview responses or articles that are 'laundered' by PR or media relations departments almost always are.)

1a) up2date - by aldousd666

Is the up2date service going to continue to work for us end users who still use RH9, or are we going to have to go Fedora treating our existing installations as defunct? I've spent quite a lot of hours configuring my systems, and I think you're going to make a lot of angry users if things change too drastically. I know a number of people who are already shunning the name red hat in favor of the other flavors.

Szulik:

up2date as shipped with Red Hat Linux 9 will continue to function against the RHN servers for up to six months after RHL9 goes out of maintenance on April 30, 2004. Fedora includes an up2date that can speak with Yum and Apt repositories and can work completely without using the RHN servers. From a sysadmin's perspective, the tool is nearly identical to what was used before; it simply pulls the packages and data from a different location. It also lets you pull both official Fedora packages as well as third party packages created by other Fedora users and developers as well as create your own repository for packages you want to distribute among your own systems.

Users continuing with RHL9 past the end of its maintenance window will be interested in the Fedora Legacy Project, a community-driven continuation of updates for RHL9 and RHL7.3.

1b) Return on RHN Entitlements? - by Anonymous Coward

I would like to consider myself a red hat advocate. It was largely based on my recommendation that 50 RHN Entitlements for updating non-enterprise version of red hat GNU/Linux. My boss has since been rubbed the wrong way when RHN failed to "work as advertised" on August 29th. The best explanation that I have gotten from red hat is that it is "the nature of SSL" that forced manual upgrades of up2date & up2date-gnome for each system. In October, red hat charged a renew fee on the 50 RHN Entitlements for another year of service. So, now that my boss has gotten the bill, he is asking what type of return on investment he should expect from May 2004 to October 2004. To make a long story short, the question is, are we being charged a full year for only 7 months of updates? If non-enterprise contracts aren't fully honored as advertised (automated updates require manual updates after Aug 28th and a full year charge only provide 7 months of updates) then how does red hat expect advocates of red hat to successfully encourage the companies that have gotten burned to pay out even more for enterprise contracts?

Szulik:

The SSL issue in August was an unfortunate result of transition inside of RHN. Although it was a significant inconvenience to our users, it was actually the result of our own tight security policies, and at no time was the security of our service at risk. Numerous steps have been taken to ensure this does not reoccur.

The entitlement renewals that occurred shortly before our recent announcements were limited and stopped when the changes were announced. Although the end of life for RHL9 was announced when RHL9 was first released, many users are in a situation with entitlements going past the end of life for Red Hat Linux. For those in this position, entitlements to both Red Hat Enterprise Linux ES and WS will be made available for the remainder of the subscriptions. in addition, discounts are available for Red Hat Enterprise Linux to any RHN customer.

2) Opportunity for small business - by salesgeek

Matthew - If you were looking for an opportunity to start a small business (size at peak $25 Million revenue, perhaps 250 employees) in the Linux world, where would you go?

Szulik:

$25M is not a small business. It's about the size when someone crazy in your organization suggests that you go public. I believe that the IT industry has increasingly adopted a transactional and services model. Differentiated service skills around Open Source software will be in demand based upon the large transition which will occur over the next 10 years as businesses transition from proprietary to commodity hardware and open source software.

3) What's next? - by Mr. Sketch

For the average person, RedHat _is_ Linux. Who do you believe will replace you as being the defacto Linux distribution for the average person?

Szulik:

The definition of average should be clear. For the 'average' reader of Slashdot, the Fedora Project is the ideal Linux distribution. For the average knowledge worker in an office setting, we believe Red Hat Enterprise Linux v.3 WS is appropriate. For the average person that needs to be able to plug in their digital camera without going into the terminal window, we think that the user's experience with any brand of Linux will be sub-par. We hope that consumer-focused technologies will thrive and mature in the Fedora Project setting. When the code is production quality, Red Hat will make them available as part of a supported distribution.

4) Server without Desktop? - by drinkypoo

One of the (many) factors leading to Microsoft dominance was that they had, from the user's perspective, essentially the same operating system on the desktop and the server, in that they ran the same software; And recently, Microsoft has provided literally the same software on desktop and server. red hat began with a general-purpose product, and then moved to an artificial separation between desktop and server as Microsoft now has, and has since moved to providing only the Server. Do you feel that this is a necessary product of the differences between open and closed source models, or is it simply the right position for red hat to take, and not the rest of the Open Source Unix community?

Szulik:

Recently we launched a statement of direction - Open Source Architecture for the enterprise. As more large customers move to distributed computing architectures, firms will want to leverage the flexibility and independence a integrated stack can create for a business. Our product line is being built through the delivery of software sold modularly. For example, our cluster suite.

5) If you could go back in time - by AftanGustur

If you could go back in time with the knowledge you have to day, and live the dot-com years for a second time. What would you change in Red Hat's business model?

Szulik:

Nothing. Three critical events occured during 1997-2000. Red Hat was able to capitalize itself for the long term. The Linux kernel continued to scale in performance and application availability with each increase in performance which helped to drive the enterprise adoption of Red Hat. These were matters of when and not if.

6) Will Red Hat become more proprietary? - by divec

One of the strengths of Red Hat has always been its emphasis on Free software. Unlike, say, SuSE, which contains significant pieces of SuSE-only infrastructure (such as YaST), Red Hat has always been more careful not to "Weld The Hood Shut". This is one reason we recommend Red Hat to customers at work.

Will we continue to see this, or will Red Hat start trying to beat the competition with proprietary add-ons?

Szulik:

No. For over 10 years Red Hat has built relationships with developers, ISVs and customers on the brand promise of delivering software based upon the GPL license in collaboration with the Open Source community. If you look back over the past 5 years, you will see the failure of companies that were building hybrid models which could not deliver the consistent value of open source code over time.

7) Diverse Hardware Support - by capt.Hij

One of the biggest issues for putting gnu/linux on the desktop is more support for hardware. I understand why Red Hat is supporting Fedora and focusing more on industrial clients, but I am concerned about the long term implications. What will Red Hat be doing to increase hardware compatibility and support? Without an official Red Hat "civilian" distribution do you feel that you will have the ability to sway hardware manufacturers to support gnu/linux?

Szulik:

3 important activites will have to take place before we see a significant increase in GPL'd hardware driver support. A large marketplace develops, customer demand and a viable supplier exists to deliver and service the integration. I'd say we are at the early stages worldwide to respond to these requirements. Increasingly we are receiving more support as compared to 24 months ago. I believe the civilian version will be filled by Fedora which will develop into a solution for many.

8) Did The Consumer Stream Make A Profit? - by reallocate

Has Red Hat's shrinkwrapped consumer-level product stream ever made a profit? To your knowledge, has SUSE or anyone else over made a profit from consumer sales?

Szulik:

Profitable yes. Was a shrink wrapped version sold at retail an economic model to grow a company? No. discounts leave a small amount of available profit. I can not speak for SuSE economics as until recently they were private.

9) personal OS choice? - by BigGerman

Which OS and desktop environments you, your colleagues and friends use every day?

thanks in advance for your honest and direct answer.

Szulik:

I have not used proprietary software for many years. I run a 5 node Linux cluster at home. I use Gnome.

10a) Education and Research Markets - by Frater

I work for a world-renowned research institution. We have ~500 Red Hat Linux systems in labs and on desktops, mostly administered by scientists and technicians rather than central IT staff -- so keeping them up to date is a challenge.

We have twice, over the past few years, attempted to contact Red Hat regarding site licensing or educational volume licensing for access to Red Hat Network. Both times the answer has been that -- unlike Sun, Microsoft, Apple, and our other OS suppliers -- Red Hat has no licensing programs for the education and science markets. For this reason, we have turned our Red Hat Linux users away from Red Hat Network and towards FreshRPMs APT [freshrpms.net] as a source of regular software updates.

With the discontinuation of the Red Hat Linux product line, we are now at an impasse. We do not expect FreshRPMs to conjure up security and bug-fix updates for a system that will no longer be supported upstream. My clients would prefer a more guaranteed solution than FreshRPMs. However, Red Hat still shows no signs of interest in the education and research market. Fedora is not an option, as we can't expect our science staff to accept major upgrades every 2-3 months -- they are science nerds, not Linux nerds.

Is there any chance that your plans for Red Hat Enterprise Linux include site- and volume-licensing oriented at the educational and research community? For if not, my colleagues and I will have a hard row to hoe -- migrating existing Red Hat Linux users to supportable distributions such as SuSE or Mandrake.

10b) Academics... - by PseudononymousCoward

Mr. Szulik,

As a professor at a Big-10 University, I now find myself in the curious situation that RedHat, for either server or workstation usage, is more expensive than Windows, owing to the terms that MS offers academia and the new licensing of RH products. Most Universities can _purchase_ Win2k3 Server for the price of one year of RHEL WS support.

Does academia constitute one more market segment that RH is no longer contesting?

We have rolled out an education plan which was priced between $25 and $50 for client and server quantity one for an annual subscription. I believe the pricing and service relationship will begin to address a void filled by the Red Hat Linux transition at an affordable price.

10c) licensing issues - by painehope

when will RedHat have a more reasonable licensing scheme? Your licensing is excellent for corporate enterprise workstations, and I realize that you are moving away from home users, but what about clusters and universities?

For example, I run Redhat across a rather large (> 4000 CPUs) cluster, and have never bothered doing more than buying a few boxed sets due to the fact that I have never been able to get a reasonable price from your sales team. Cluster support tends to be more like dealing w/ a single machine, since the hardware is generational (if you add 512 CPUs to the system, their hardware is going to be exactly the same if you ordered it that way). Why should I pay a license for each machine, when I can just get a license for one that is having the same problem as the others (for example, a bizarre problem we had w/ the eepro100 driver + PVM - and yes, I know PVM is generally used for > 1 machine, but technically I probably could have addressed the support problem w/ 1 license). I wouldn't have a problem buying cluster support if you had a decent sliding scale (ex. : 512 nodes @ $50/node, 1024 nodes @ $35/node, etc.). And of course, have a caching update server for the site.

And for universities: if you want brand recognition, try offering site licenses or educational discounts. Don't count on all CS/EE students to be clued in enough to install Fedora on their laptop and then debug any problems that come up. Offer a site-wide license to all students for $50k, or a department for $10k, or something like that. That would probably give you a lot of name recognition in the future. You already offer site licenses for corporations, right?

So when will RedHat come up w/ some decent licensing schemes for those environments?

Szulik:

Painhope, my view of reasonable and your view of reasonable might be different. And I would like to take you up on your offer. Send me an e-mail and we will take you up on your offer. Keep in mind that we do not sell licenses. We sell subscriptions where the value of the bits are integrated with service levels. I believe our educational subscription plan will be seen as a good solution to opportunities like yours. And you are correct, most student computing activities must be supported by campus IT to get plugged into the campus network. Site license for $50k. For many public schools and university, this is a large sum.

User friendliness

For the average person that needs to be able to plug in their digital camera without going into the terminal window, we think that the user's experience with any brand of Linux will be sub-par.

Why, oh why must it be so?

Why is it so hard to have real user-friendliness in Linux?

Re:User friendliness

It's a catch-22 - the average Joe on the street doesn't want to touch Linux until it's friendly, but companies don't want to invest the time and money to get their goods working on Linux unless the user base is big enough to give them a good return on their investment.

Hell, I can see why the average person won't touch Linux - I bought a Radeon 9200 dual head video card 2 weeks ago, and after off-and-on screwing with it, I still haven't gotten both heads working at once, excluding cloned mode. Some soccer mom isn't going to want to be dicking with XF86Config while her kids are bitching about wanting to play Pokemon...she's going to want to boot into Windows, pop in the CD that came with the product, and have it work.

Re:User friendliness

My wife is the classic soccer mom, van and all, and has been modifying config files -- DOS then and Linux now -- when you were still wearing plastic pants.

And does the average soccer mom have the same abilities as your wife? Survey says...nope. There are always going to be exceptions to the genealization, but they very rarely make a point invalid. The average person, be it soccer mom or not, isn't going to want to screw with their computer to get it working.

Unlike you, she's smart enough to use the right tools for the job.

And unlike you, I understand the concept of an example. :-)

Re:User friendliness

My wife is the classic soccer mom, van and all, and has been modifying config files

By definition, this is not a classic soccer mom.

If you don't believe me, get some of her fellow moms from soccer practice and give a command line and ask them to install Slackware.

Or give them a DOS prompt and ask them to do anything.

Or give them Windows and ask them to go download, install, and run Mozilla.

This will educate you in ways those of us who didn't get a geek with breasts for a wife cannot.

Re:User friendliness

No one wants to work on the little "niceities" that a user-oriented OS such as Windows or Mac OS X has. OSS developers want to work on the "cool" projects, that and for us, functionality is relative to the user. OSS developers tend to write software for developers, not your average Joe Sixpack.

Fine by me, though. I'm not afraid of the command line. :)

Re:User friendliness

Many camera owners that use a Windows PC like use the software that came with the camera to download images -- thus far, there has been very limited support from manufacturers for linux. It is really a shame that someone like Canon does not create some linux software and include it with their cameras. Similar issues exist for other hardware, although exceptions exist such as USB mouse.

I have a PowerShot A70 that I use under linux with mild success -- this usually requries cycling the power on the camer a few times before I can get the images. This generally does not bother me, but I have a feeling my mom would be less than excited to do this.

Re:User friendliness

Almost all Canon digital cameras work very well with Linux; you plug in the USB cable and go (provided that you've installed a gphoto2-based app such as gtkam or kamera). Some other brands don't work as well.

Digital cameras are actually an area where Linux does quite well (for some other devices it's a different story), and the Linux situation even has some advantages over the Windows situation. Consider the digital camera owner who visits a friend's house, and who brought her camera's USB cable. Can she upload her pictures to her friend's computer? If the friend runs Windows, probably not, if the friend has a different brand of digital camera. If the friend runs Linux, and it's one of the hundreds of supported camera models, it just works.

The Windows user could bring her software along as well and install it on the friend's computer, but this would violate the EULA and potentially subject her to prosecution!

I don't want digital camera makers to include binary-only drivers and put us in the same box that Windows users are in. Instead, they should document the protocol used on the USB cable, so that the gphoto2 people can add support (gphoto2, despite the "g", is used by both KDE and Gnome apps to access digital cameras) . Better yet, they can submit source code to the gphoto2 people.

Hardware support will help a great deal

"Real user friendliness" with respect to mass-market peripherals requires the support of the manufacturer of each peripheral. Until we see a SANE driver next to the TWAIN driver on the CD that comes with a scanner or camera, there is little that the community can do, especially if a manufacturer refuses to disclose its peripherals' wire protocols to the community.

Re:User friendliness

[Note: I am using any Linux commands as examples from personal experience, I am not taking the effort to ensure that the linux terminology I'm using is perfectly correct, because it's supposed to be an example from the lay-perspective anyway, so, no nitpicking my mount points, 'cause I don't care]

Just because it's not quite there yet, more than anything. I don't think anyone is claiming that Linux will never, ever, EVER be suitable for the mass market, but, well, damnit, it isn't yet. The console commands are confusing to those without experience with them, and they aren't always perfectly documented from the layman's perspective (something like "-con350 will shift my R-variable? Why the hell would I want to do that? What IS that?").

From someone without computer experience, what the hell is "mounting" a hard drive?

Where do you put your pictures? Not in "C:\pictures", but in "/mnt/users/username/home/pictuers" or somesuch.

Hey, how about "How does someone with no Linux experience install things?" - They don't. "That program I downloaded doesn't have an install file!"

It's a good OS. I've got two flavors triple booting on my machine, so I can learn it. It's absolutely great for coding. It's got solid support for the basics I want to use my computer for, and, with effort, I can always get whatever else I need to work- With effort, because I'm new to Linux, and because, thank god it's possible because I'm computer literate enough to know damn well that it will work *somehow*.

Re:User friendliness

Console commands aren't normally needed. That isn't to say that you can do everything without them, but the majority of Linux users would rather have it that way. It allows for flexibility. Unless people with complaints start writing GUI-based apps that change that, there will never be a change... And the geeks won't care.

Mounting a harddrive isn't necessary to do by hand with Automount. Most "user-friendly" distributions use it. Windows also "mounts devices" but the way the file system structure works is different. Drives (most often) get letter assingments, while UNIX devices get attached (or mounted) to a directory on the tree. Either way, with automount, it doesn't matter. Many distributions detect when a new drive is installed. From there, it configured the fstab for the device so it mounts on bootup.

In comparison to Windows machines, in which you have something like "C:\My Documents\Username\My Pictures", Linux isn't really any different. A lot of programs default to "/home/username/" for file storage. Is it hard to add a "Documents" folder with Nautilus? I use one on my machine, and it works rather well. Nautilus (and other file managers) default to opening the user's home directory... Not some directory in /mnt as mentioned. It would make no sense for a program to default to "/mnt/users/username/home/pictures".

Yes. Linux has install files. They are generally with the extension SH, BIN or RUN. Theyrun the same way that Windows EXE files work. Most commercial Linux games come this way. They are self extracting and have graphical installers in the majority of situations. The difficult part is that you often have to launch them from a shell window. Programs like Nautilus tend to load SH script files up as text files, unless you teach it to work differently. Not sure about the KDE side of things though.

Re:User friendliness

"real user-friendliness" and "automated support of recent consumer hardware" are very different things. a lot of people find the GNOME and KDE desktop environments to be very user-friendly, as far as the current state of the Linux application base and hardware support go.

I, for one, find it very friendly that I can install and upgrade my OS for free, and that when I do so, it includes all of the things I need (web and internet applications, media software, graphics software, office and productivity) ... with the base installation.

as mr. szulik says, improved hardware support will depend on hardware manufacturers' cooperation, for the most part, and thankfully that has been improving of late.

Re:User friendliness

Personally, I think it's because real user friendliness is a QA issue. And not just QA of software but of a whole system, how all the modules and packages work together.

That's something that most "geek" projects have a hard time handling. It's just too large of a problem. They don't have the time or resources to QA their's and everyone else's work.

And as I think about it, it might be tough for a large company too. Take all that work and QA it, turn it into something useful. Hard, hard, and that's what most people like doing, right, is fixing some one else's work? Not.

So I was going to try to give you an answer but no I've talked myself out of it. Too big for geeks, and too big for companies too. I think perhaps we need a paradigm shift. Something that allows individual geek projects to work together better. Something like Extreme Distributed Software Engineering. So that the QA of an integrater for a distro is much smaller and easier.

Hmm, maybe this is a role that the DLC [desktoplin...ortium.org] or OSDL [osdl.org] could play.

Re:All hail Microsoft, the Marketing Genious.

Because like it or not Exchange and SQL2000 are far more advanced than anything Linux has to offer.

More advanced than Oracle ?

And now that their are interfaces, people are going to try and put out applications in *nix with Mono. Only forcing companies in the long run to buy Microsoft Servers and might as well buy Desktops too, because .Net will be .Net where ever you go.

Personally, I never plan to develop with mono. Gtk+ works well enough for me. And I don't know of any other Linux developers who plan on using mono either.

Besides, if RH are pulling out of the desktop market, that just leaves more room for the other distro's.

painehope...

Painhope, my view of reasonable and your view of reasonable might be different. And I would like to take you up on your offer. Send me an e-mail and we will take you up on your offer. Keep in mind that we do not sell licenses. We sell subscriptions where the value of the bits are integrated with service levels.

How much is Redhat going to ask for per CPU? painehope gave an example of $50 per CPU for a 512 node machine, $35 per CPU for a 1024 CPU machine, etc. How much is his 4000 node machine going to cost?

All in all, a good interview. Szulik even runs a Linux cluster at home! very nice.

Moderation Limitation?

Perhaps for Interview questions, the moderation system might be modified to remove the limit of +5?

Re:Moderation Limitation?

Heh.. Just seeing the question

Are you waiting for Microsoft to buy you out? (Score:45, Flamebait)

would make my day

He skipped the Edu questions...

I like how he totally breezed over the educational institution questions, which were of particular interest to me as a net admin at a big 10 univ who also has RedHat machines to maintain that will soon be completely unsupported. I need to maintain a lab of computers that dual-boot to Linux and WinXP, and RH9 is the best desktop Linux solution out there currently. I don't have the budget to get the Enterprise edition, so what am I supposed to do when the support runs out?

-Mike

Odd response to questions 10a/b/c

It seems to me than rather addressing the issue and assuring customers that educational instutution strategy and licensing will be prioritized, Szulik goes on to pitch a deal (on slashdot, no less) to the person asking the question.

And I would like to take you up on your offer. Send me an e-mail and we will take you up on your offer.

I'm not amused. Clearly, Red Hat isn't doing enough to accommodate educational facilities with discounted volume licensing.

Digital Camera Comment

Oddly enough, my main home machine is a RedHat 8 Athlon 2100+

My wife has a digital camera, and it Just Works. Plug in the camera, start "Digital Camera Tool" from the Gnome start menu, download pics. No shell window required.

My non-techie wife has no problems with it at all.

Now getting the Espon C82 printer to print photos with any sort of colour fidelity was a weekend of build-CUPS-from-scratch HELL - but the camera was a no-brainer.

The RedHat desktop user experience is nowhere near as bad as it is made out to be.

DG

perfect English

[U]nlike many CEO responses, Szulik's answers to the selected questions are his own, not PR-generated. (One clue is that they are not in perfect English, as interview responses or articles that are 'laundered' by PR or media relations departments almost always are.

Ah. If you speak well, you must be insincere?

Time travel?

If you could go back in time with the knowledge you have to day, and live the dot-com years for a second time.

Darl McBride, you have been scheduled for termination.

A bit obsequious, no?

And unlike many CEO responses, Szulik's answers to the selected questions are his own, not PR-generated. (One clue is that they are not in perfect English, as interview responses or articles that are 'laundered' by PR or media relations departments almost always are.)

That seems a bit obsequious to me given the actual information provided. Except for a few tidbits about discounts for continuing support customers, was there a single piece of real information in there?

It's not like he was offering answers like Marcelo Tosatti's -- the fawning "No marketing here!" comment seemed a bit excessive.

RHEL/Pro/Academic Product Differentiation

But my main question wasn't answered. I wanted to know about the shadiness surrounding the Red Hat Enterprise, Professional Workstation and Academic versions. The latter two seem to be rebadged/repackaged versions of Enterprise WS.

My original comment from the Q&A article.

Why isn't Red Hat actively marketing their Professional Workstation Product? [redhat.com] Apparently, this is a newly-released offering that hasn't been receiving much attention. It's odd, because it's not even displayed prominently on their site [redhat.com].

However, a Google cache of the page [216.239.41.104] shows the relationship of Professional Workstation to the rest of the RHEL line.

The Red Hat Professional Workstation isn't available online, or through Red Hat, but through a few selected retail channels. Buy.com has it for $82.57 [buy.com], which includes one year of up2date service. It's the same product as Red Hat Enterprise Workstation. I purchased it from my local Microcenter for $99. Here's the RPM list [ewilts.org].

It looks like this product was a last-minute addition.... Apparently, it's not crippled or relabeled.

Given my previous rants on Slashdot [slashdot.org] about the Red Hat shadiness, this looks like a good option.

Even more interesting is the fact that Red Hat didn't put much effort into product differentiation with this Professional Workstation product. I opened the box and the CDs were labeled "Red Hat Enterprise Linux WS". Well, only the first CD was labeled as such. The other CDs are identical to the Red Hat Enterprise AS/ES offering and include the same RPMS/SRPMS. SRPMS build cleanly in every test case I tried. So, buying this and using Enterprise 3.0 SRPMS for future updates is entirely possible. The same RHEL patched 2.4.21 kernel is there, too. Nifty.

Another issues that bugged me about the Red Hat Enterprise Linux move was the poor upgrade path. Reinstalling the OS on production servers that are running Red Hat 7.x or 8 ain't pretty. So, my final test with the Professional Workstation was prompted by a half-page paragraph in the manual that came with the box set.... It stated that in-place OS upgrades were only available for Red Hat Enterprise 2.1 -> Red Hat Enterprise 3.0 systems (via "linux update" at boot)...... however, you have the option of booting the install CD with "linux updateany" to relax the restriction "in case your /etc/issue file is damaged". Hmm.... No version-checking, eh? So I performed a test in-place upgrade on an existing Red Hat 8.0-equipped Proliant server...... It totally worked without a hitch!

This, along with the education and bulk-pricing deals leads me to believe that the Red Hat marketing department is working hard to appeal to the people it alienated with its announcements over the past few weeks. But it may not be enough. How can enyone plan for the future when Red Hat seems to be a moving target? We'll see what happens come December 31.

Wow

Painhope, my view of reasonable and your view of reasonable might be different.

I must say, I am floored. If RedHat's view of reasonable differs from its customer's view of reasonable (which it obviously does), then this is going to be a disaster.

As a consultant and long time RedHat user who has brought RedHat into many companies with no Linux presence, I can no longer recommend them. RedHat's primary advantage was its low cost for a fully supported product. Now, that advantage no longer exists.

Sad too, because RedHat was really starting to gain some brand recognition. Now, it's going to be known as the Linux that's too expensive to use.

Re:Wow

Are you kidding? painehope wanted RedHat to support a 4000 CPU cluster for the price of a few lousy RedHat Linux boxed sets. That's just insane.

RedHat's support is still far less expensive than any commercial competitor. With Windows support is also separate from licensing. The difference is tha t with RedHat licensing itself is free.

If you want you can switch to SuSE (pretty much the only other Linux choice that is supported by a commercial company), but don't expect to pay any less for actual support (not to mention the fact that YaST2 is not Free Software).

Personally, I use Debian, as I am not interested in commercial support, but I can't blame RedHat for trying to get a fair price for their support. Engineers don't come cheap.

Camera ?

For the average person that needs to be able to plug in their digital camera without going into the terminal window, we think that the user's experience with any brand of Linux will be sub-par.

I guess he has yet to try SuperMount [sourceforge.net]

Sunny Dubey

Licensing, support and updates: Debian

It seems like the three main issues that I identified are:
Licensing, particularly for educational institutions,
Support, especially for the soon-deprecated RH9 series, and
Updates, and the continuation of the up2date network.

Many of the users of Red Hat seem understandably confused and upset about the direction that the company is taking. I would like to, humbly, suggest that none of these issues are pertinent to the Debian distribution. I would personally encourage users in a situation where they feel tramelled to do some research in this respect.

I think it would be inappropriate, in the context of posting to this interview, for me to suggest Debian as an alternative to Red Hat Enterprise edition. However, I do believe it to be a substantial alternative to the soon defunct consumer Red Hat series. In time, Fedora may also be a valid alternative, but at the moment its capacity to act as a valuable, low risk distribution has not been substantiated.

Still don't know the answer...

The questions were good, but I think that something is missing...

What about stable releases for fedora?? Since it's gonna be the "bleeding-edge" will it ever be a clear distinction between stable and development? Will the security bugs be worked out and patches made available or will people need to upgrade all the time?

I'm not a red hat or fedora user (long live slack!) but I've got some friends that barelly know their way around RHL. Putting something full of holes in their hands will only frustrate them. Well, I guess I'll have to talk them into something like Mandrake or Debian.

Redhat doesn't sell licences?

Okay, perhaps someone here can fill me in. I'm not too au fait with the business models used by these companies, and just stick to my pet distros.. but he said this:

Keep in mind that we do not sell licenses. We sell subscriptions where the value of the bits are integrated with service levels.

This sounds fair, considering Redhat is just a bundle of open source software, with a few pieces of free, but closed source, technology.

But does this mean I could legally get a copy of RedHat Enterprise Server, and install it on as many machines as I want? That is, I don't pay for any support or ongoing upgrades, but I get the benefit of the new Redhat product. Then, for support and upgrades, there's always the community.. or, as I'm sure will happen, a 'free' community effort to keep RES patched against the major problems will spring up.

Answers, paraphrased

I wasn't very impressed. Maybe I'm just bitter about having to migrate from RH9 to "something else"... which will probably be Debian rather than Fedora (due to Fedora's very short maintainance schedule).

Here's roughly how I read his answers:

1) Your fully paid RHN subscriptions for RH9 will be worthless. You can't have a refund, but you can pay even more for a (discounted but not complementary) upgrade to the enterprise version, to keep using your already-paid RHN entitlements. And yes, we can almost admit the SSL problem that broke RHN was our fault, but rather than appologizing, it was actually due to our excellent security policy.

2) No actual recommendation for small business. Dodge the question by babbling about what a small business is, and something about "Differentiated services skills" during the transition from proprietary to open source deployment. Yeah, sure Roblimo, that's not PR speak! I've got a nice bridge for sale too. Wanna meet in Brooklin to see it?

3) Slashdot readers should be content with Fedora, but everyone who works for a living should pay for various versions of Redhat Enterprise Linux. Non-technical home users should use Microsoft.

4) Redhat decided only to focus on enterprise. Hint that WS is meant to be a client, but ultimately dodge question about advantage of offering "full package" including both desktop and client.

5) Redhat has never made a mistake, even in the dot-com days.

6) Promise to stay with GPL and collaborate with open source community.

7) Hardware support won't come until a large user base demands it. Fedora is supposed to build that user base. Dodge the question about discontinuance of RH9 affecting growth of hardware demand.

8) Shrink wrapped product didn't make enough money and couldn't grow (but no admission it was unprofitable, despite the well-known fact that Redhat was always in the red all those years).

9) Matt user linux and gnome at home.

10) Call us regarding your educational discount... because we won't say anything specific in public.

Answers less than clear? Questions poor?

I don't know if it's only me but I found his answers to the less than good questions poor. Apart from the questions about up2date and the educational issues (10a/b/c) , I found one big question missing that had been modded up to 5, and that is the question about what will happen to RedHat's image due to leaving the non commercial space.

Many people were wondering why RedHat did this, since the non commercial space is where most people got to know RH in the first place. My personal reaction to this is that I went out and bought a Mandrake subscription, as I felt that RH had sort of "betrayed" it's most loyal users. I see no real difference between Fedora and Gentoo and I felt that the one company left supporting non commercial users, Mandrake, was worth supporting. I see an image problem for RH in gaining new geek advocacy in future. It remains to be seen what becomes of SuSE's non commercial efforts.

As for the questions about educational institutions, I found his answers very poor. Why did painhope have to wait this long to get a reply? Why were RedHat sales teams so ignorant of educational pricing from Microsoft that they neglected customers like this, until it got posted on /. for the world to see?

To me, It sounds like RH has a very disconnected view of some important issues in the real world. Number one is lack of perception from customers' point of view and number two is an incredible lack of perspective and proactive action on RH's part: If the desktop was profitable, and considering the fact that this was RH's public image, then why not keep it for simple reasons of good PR. If there are so many driver issues (web cams, digital cameras etc) then why on earth didn't RH simply approach some companies in order to get a Linux effort started with those companies? The way he says it, it sounds as if he's simply too bloody lazy and disinterested in actually listening to customers.

Average ha!

For the 'average' reader of Slashdot, the Fedora Project is the ideal Linux distribution.

The average slashdot reader runs windows and is a little technical. They need something that can work long term to play with when they want. They don't want something they have to upgrade every three months!!! I know enough to install Debian which is the hard part. The rest is easy. It just runs and is easy to update over a _long_ period of time without fuss. I have been running stable on some of my boxes for a number of years. I am not a kernel hacker, I enjoy just playing around in Linux and don't care having to get everything up and working every three months! Back when I ran windows I reloaded it every one-three months and I remember just how much of a pain in the ass it was, so much so that I tried Linux out... Especially when you make it painfully clear that Fedora is nothing more then a beta test for the enterprise beta/release don't fill me with happy thoughts of productivity. And the idea of having multiple rpm repositories gives me the willies. In the debian world most of the package are in one place (or heading to it). Because of that they can be made sure to be compatiable and well intigrated (or hell get this... just work!) This is a good thing. What happens when repository #8 in on your box just goes off line because the guy that was hosting it moved? I tried out the apt-rpm and found myself having to spend quite a lot of extra time updating because the servers were always busy. I have never gotten a busy reply from the debian repositories. Again they just are there and work. Just wait till the first time someone updates a Fedora box and repository 1 has kernel 2.6 and repostitory has kernel 2.4 and you grab parts from each and then turn on your computer the next morning. NO THANKS! You average slashdot user wants somthing that just works and continues to work, they don't care to re-install every three months or have to worry about which pagages work with what other packages.

-Benjamin Meyer

The real cost of updates and fixes.

I've been informally tracking conversation threads about Fedora Core. It seems to me that a lot of people are dancing around the question of how much security and bugfix updates are really worth. Apparently many people rely on Red Hat's security and bugfixes to where they will be missed if absent. But at the same time, people don't value them enough to pay (virtually anyone) for them, hence they would rather switch distributions than pay for the software improvements they've come to rely on.

Perhaps this is the watershed event that makes people aware of what a service economy looks like when people have to deal with something close to honest pricing--the end of getting Red Hat's widely-appreciated labor at no charge.

Yeah Right

"One clue is that they are not in perfect English, as interview responses or articles that are 'laundered' by PR or media relations departments almost always are"

And how would the editors at slashdot know what perfect english looks like?

Academic Pricing :: goodbye redhat

After seeing that non-response to the academic pricing, I was annoyed enough to call redhat up to see what is going on. They have not finalized the pricing but it is going to be $25->$30 a year per machine w/ some minumum number of machines. They also will offer a "proxy server" for $2500 so you can roll your own updates.

(1) The proxy server idea is stupid since apt is ported over to redhat. If we know how to setup a ftp server we automagically get a "proxy" server. Unless they are not going to let us grab the updated rpms (from
our $25 per node licensed "up2dates") and copy it to the server which if they dont as far as I can tell violates the gpl (unless they do an "endrun" by providing rpm's w/ "trademarked" pictures in it).

(2) $25 a machine/year is more expensive than windows for updates! Yes redhat is a smaller company BUT at microsoft is writing their own gui, writing their own kernel, writing their own patches,
writing their own driver specs. Redhat is NOT doing these things on their own (but to be fair they are contributing but having per machine licensing is a trick worthy of SCO not a linux company).

(3) We are in a physics dept and run "oscared" images for a smallish beowulf cluster (50 dual nodes) and have two more beowulf clusters and one public access workstation cluster in our department alone. It makes *no* sense to pay $25 a node for this since we "automirror" when things go awry. I personally will move over to debian (or wait and see what fermilabs etc is going to do). This redhat fiasco is a fiasco. Using "trademarked" pictures to do an endrun around the gpl to get "per processor" licensing is an end run around the GPL and ought to be treated as such. (This is, as far as I can tell, what they are doing w/ "enterprise" redhat to prevent me from buying one "enterprise" redhat and apt-proxying my other nodes). Feel free to correct me if I am mistaken.

best regards,
-bloo

Selling out

I've gone out of my way over the years to buy Red Hat box sets and up2date subscriptions. I wanted to give money to Red Hat because they were pretty cool in the early days when they seemed to have a long view and had what it would take to make Linux successful. After recent events and reading Matthew's BS I'm afraid I made a mistake. For what ever reason the executives at Red Hat have mutated in to the same kind of slimeballs that populate companies the world over and they certanily don't "get it" anymore. Its a really great idea to dump your flagship producton, and customers, on the backs of a bunch of unpaid volunteers, while you try to cash in on the peices you can charge an arm and a leg for. Red Hat's strategy of making a product based on free labor worked when they were making an affordable and convenient package. Now that they are orphaning a big segment of their loyal customer base and still leveraging free labor but are now charging rates rivaling Microsoft's and are starting to look just as slimy and unappealing. It was a really great idea to sign a bunch of people for up2date and then immediately kick the chair out from under it leaving paying customers hanging.

I imagine this can be attributed to the disease that inflicts every company after it goes public. They stop prioritizing making good products and cultivating happy customers who in turn give them money with joy in their hearts because they like the product.

Instead, like most public companies, the only people they start caring about are the analysts sitting in Wall Street and the one and only priority is making the revenue projections each quarter so the stock price goes up and they get rich when they cash in the options. Just being profitable isn't good enough either. MUST GROW FAST AND CONSTANTLY whether its sound business or not. Customers, rather than being the top priority, turn in to a necessary evil who must be constantly milk for cash and they must be constantly manipulated.

Priority #1, must get customers to sign up for subscriptions. Just selling good software is too unpredictable. If we screw the pooch and a new release sucks people don't buy it, we miss our numbers and Wall Street is unhappy. If we make customers pay us a constant amount of money each year then we ALWAYS make our numbers even if our product sucks sometimes.

Priority #2, a key component of subscriptions is support. But damnit support is expensive. Must cut support costs. Lets hire a bunch of people in India who are dirt cheap. Nothing wrong with that if they actually know what they are doing. The problem is they are usually hired iike cattle and handed a bunch of preprinted FAQ's. As long as the customers question is precisely answered on the FAQ service is great, unfortunately the FAQ's only work half the time and the rest of the time your support staff exercises their one true skill, using the buttons on their phone to constantly forward or put on hold anyone who has an actual problem until they eventually give up and hang up.

Priority #3, make sure all your competitors are also publicly traded and also implement Priority #1 and #2 so they suck just as bad as you do so customers are left choosing between the lesser evils and will pay you even though your company's products have started to suck. Thats what competition is all about. Everybody competes to be equally shitty.

After some consideration I've deduced that Capitalism was an interesting experiment but its reached the point its flaws are starting to far outweigh its benefits. Fact is its become 100% about overpaid and unscrupulous execs striving to make as much money as possible as easily as possible. Screwing labor and customers is job 1. Those with ethics and interested in producing a good product at a fair price need not apply. All of Capitalism's competitors have also proven to suck so maybe we should go back to the drawing board and try to come up with a economic system where people are actually rewarded based on the merit of their work.

Re:APT?

He answered that in the first question.

"Users continuing with RHL9 past the end of its maintenance window will be interested in the Fedora Legacy Project, a community-driven continuation of updates for RHL9 and RHL7.3."

Based on the talent involved with the Fedora project I see no reason as to why updates won't work well and in a timely manner. To doubt that is to cast doubt on every single other community Linux project such as Debian, Slackware, Gentoo et al. Ther are a lot of users in your particular situation. Nobody has a desire to make things more confusing. I'm absolutely positive that by the time official support ends next year that a great replacement will be up and running for the countless Red Hat users out there. Red Hat 7.x through 9 is likely to be supported for years to come. So no, hopefully tarballs won't be in your future.