Enterprise FOSS Adoption Beyond Linux Servers?

An anonymous reader writes "I am working with a couple of large companies that are purchasing web and collaboration software stacks from Microsoft, IBM and others. These are for thousands of end users and are (supposedly) ready for multiple data center deployment and other big-corp requirements. I have suggested some open source alternatives such as Liferay and Drupal, and the technical people are interested but management types are not. They have given a few reasons, such as concerns over supportability and enterprise-readiness, but my feeling is that they are being won over by FUD from large vendors and the fact that most corps do not have significant deployments of FOSS technologies beyond Linux yet. All this seems to be in line with a survey on Web-app servers by OpenLogic. So my questions are: How have you persuaded larger enterprises to adopt server-side OSS, beyond server-room Linux and a couple of demo JBoss boxes under someone's desk? And which products are truly ready for enterprise-scale deployment?"

Comments

-Enterprise

[MrEricSir]

Could someone re-write this story without the buzzword "enterprise" substituting for the actual requirements?

Until then, I will have to mod this down.

Re:-Enterprise

[JWSmythe]

    You just got out of Microsoft school, didn't you?

    In the real world, the majority of servers are *nix based, with the majority of those being Linux. You'll find them all over the place.

    Yes, you'll need to learn the CLI to do it right. Playing point and click just doesn't cut it in the higher levels. Even in the higher levels of Microsoft stuff, you'll need to know how to use their CLI, except it's not well documented, and a quick Google search won't tell you all the answers.

    Wait until you have to start programming. Don't worry, if you get beyond help desk support for your local ISP, telling people how to renew their DHCP lease, you'll have to (oh my gosh) actually type things. Since you're probably unaware, the nifty point and click programs were actually written out and compiled. They didn't just start life as pretty interfaces. When you start scripting (batch, VB, Perl, PHP, or whatever) you'll live in the CLI. That is, unless you live on crutches provided to you by others.

    I'm a *nix/Linux admin. I get pulled into the Windows arena on occasion. Because I'm really good at what I do, it's assumed I'm good at anything. The truth is, I'll figure it out faster than most people, which is why they call me. Once I had to add several hundred new sites to an IIS web server. They were pointing and clicking, and wondering why the occasional one didn't work (you missed a click). I wiped out the 10 sites that they had done by hand, and scripted the whole thing. My script took less than 20 minutes to write, and less than a minute to execute. It would have taken them days to get all the sites entered and fixed, and even still, customers would have called complaining because particular check boxes weren't clicked when they should have been.

    Linux and open source are in the enterprise, and they're going to stay. They are the future, and Microsoft is struggling to keep up. But hey, MS is all you know, it's what you learned in your tech school, so you could get your MSCE, and now you hang it proudly in your cube at your call center. Congratulations. If you want to succeed, pick up some more skills. Linux, Solaris, and AIX are a start. MySQL, and Oracle, Apache are good too. Pick up Perl, PHP, shell scripting, and maybe get some decent exposure to C*. Go get your Cisco cert too. Once you're there, then you're allowed to play with the big boys. Until then, sush up and answer your support calls from housewives who can't figure out what the mouse is. Don't forget those winning Microsoft skills you picked up. Once you've shown that you are great at what you do, you'll still be asked to fix office computers because they have malware or some mysterious crashing problem.

Re:-Enterprise

[DuckDodgers]

I think every solution should always be given a fair consideration. But when you factor all features, Microsoft and many other proprietary vendors enter the arena at a major disadvantage: licensing costs, software use restrictions, a dis-incentive for the vendor to make migrating to other vendors easy, a dis-incentive to make their software compatible with other vendors, and of course no ability to review or fork the source code yourself.

I never have to worry about paying an additional software licensing fee to a vendor when my PostgreSQL database passes the 10 GB boundary, or when I add another server on the domain, or when I install an extra CPU in a server. I never have to worry about being unable to buy an additional new copy of my Linux distribution. I don't violate any terms of use when I post performance comparisons or feature complaints or any other comment about the products.

Now on the bright side, I think open source software has become so good partly as a reaction to the good moves by proprietary vendors. OpenOffice plays catch-up to Microsoft Office. The various open source VMs play catch-up to VMware and such. PostgreSQL keeps racing to try and match Oracle, DB2, and SQL Server. But the gaps are getting very narrow, in some cases open source has a clear lead, and the open source licensing advantages are a very strong argument all by themselves.

Use the big vendors to assist

[xzvf]

I work for IBM, but don't speak for them in an official capacity. Open source is customer driven and not vendor driven. There is little incentive for anyone outside your company to push open source software because it reduces their profit. Ask your vendors to come up with solutions that use open alternatives, otherwise they are just going to push what makes them money. Software margins are high and ISV's are bribed to push it. I think MS gives 6% kickback to vendors that sells a license, which is a revenue stream lost when open source is used. Ask your vendors to present an open alternative alongside their proprietary ones. Same support that management demands, but less risk.

Look in the mirror

[mcrbids]

To evaluate the success of your recommendations, take a look in the mirror. What's your credibility to suggest anything at all when you have to come to (of all places) Slashdot for advice?

Large corps have lots at stake, and they really, really, REALLY are terrified of any solutions that aren't basically guaranteed to work by large, trusted vendors. Stuff that they consider to be a competitive advantage will be enshrouded in mystery while everything else will be outsourced to the most commodity vendor.

Now, compare 'Drupal' to 'Microsoft'. Maybe everybody HERE knows how painful it can be to get MS stuff to work, but nobody is going to be fired for saying MS because it's the biggest commodity vendor in the software space.

Look in the mirror: are you trusted there? When you are fired, who is MEGACORP going to go to when there's a problem?

These questions are being answered by PEOPLE who are afraid that if they make a risky decision, they will suffer the consequences. (get fired/sued/whatever) To sell your OSS solution you have to that there's no/little risk in going with it.

Good luck.

Re:Look in the mirror

[Timothy Brownawell]

What's your credibility to suggest anything at all when you have to come to (of all places) Slashdot for advice?

Presumably better than if he was the type to pretend he knows everything.

Large corps have lots at stake, and they really, really, REALLY are terrified of any solutions that aren't basically guaranteed to work by large, trusted vendors.

Is this a rational fear? It probably is for hardware, where the big vendor can overnight replace the entire system for you after a rat eats it, but what about software where the failure causes are different? How does responsiveness and the effectiveness of that response compare between the various guarantees? How often is this actually needed?

Now, compare 'Drupal' to 'Microsoft'. Maybe everybody HERE knows how painful it can be to get MS stuff to work, but nobody is going to be fired for saying MS because it's the biggest commodity vendor in the software space.

isn't this essentially the classic definition of FUD ("nobody ever got fired for buying IBM equipment")?

To sell your OSS solution you have to that there's no/little risk in going with it.

Or that the benefits outweigh the risk, else why would pretty much everyone run Windows instead of something that people don't bother to write viruses for?

Re:Look in the mirror

[kimvette]

Large corps have lots at stake, and they really, really, REALLY are terrified of any solutions that aren't basically guaranteed to work by large, trusted vendors.

Aside from hardware (game controllers, mice/mouses[?], keyboards, etc.) what does Microsoft guarantee to work? I have read their EULAs. Heck, I even worked second-tier Windows support back in the day. They expressly disclaim all warranties, stated or implied. There _is_no_guarantee_ that Windows or Microsoft Office will work for any purpose. They do not guarantee that it will work, and they certainly won't guarantee that Microsoft Excel can properly add 2+2.

With all warranties expressly disclaimed, HOW does "REALLY are terrified of any solutions that aren't basically guaranteed to work by large, trusted vendors" make expensive proprietary software a better choice than free/open source solutions? The industry entrenched around the theory that you need it, and you will NOT take your mouth off the teats of Microsoft and you will need expensive training and "maintenance."

Finger-pointing? What happens when a proprietary product reaches EOL and support is required? Many companies require you to purchase the new product even before you can purchase the support incident - if they will even support the old version at all. Who fixes the product then? If you need data recovered, it takes someone deciphering the data format with a hex editor, or trying to make heads and tails of a closed-source vendor's idea of a database schema.

When an open-source product loses its backing (project is abandoned, the company which created it is sold or closes, or whatever) who can fix it? Whatever developer you can find who knows the language the product was coded in. Worst case you'll still have access to your data and can migrate it to something else, but in most cases you can get the defect fixed and move on in life and get back to doing your real work.

When looking at it objectively:

Which is the bigger risk?
Which is the safer bet?

You might argue that Microsoft is stable and isn't going anywhere soon, but on the other hand, all you bought was 20 seats of office (or "pirated" (arrrgh!) one across 20 workstations) and to a company with $100 billion in the bank, your threat to go elsewhere if they don't fix your bug in $f00, it's less than the buzzing of a mosquito. It's not even head lice to them. They couldn't care less because a) they already have your money b) you're too small to give a squat about and c) you're ("you" in this hypothetical situation, not "you" specifically) stupid enough to keep buying their product even when they do not fix their bugs.

So, the bug will not be fixed, and you still will pay for the product. That is just how life is. However, F/OSS would have given you the software for free (BONUS!) and you would have been able to get the bug fixed. Now, it is true perhaps that fixing the bug might cost more (if you had to hire a developer to fix it for you) than Microsoft Office would have cost you, but on the other hand, the fact remains that you could fix it and gain access to your data and get on with making a living.

Now, in an "enterprise" situation I would think that in a situation where there is no warranty, and there is an option costing millions with limited hardware support and a limited lifetime and risk of lawsuits in the event of "license" "violations" and there is a free option where the support is JUST AS GOOD, if not better, supports more server-grade hardware, there is NO risk of per-seat "license" "violations" AND the source is available so you know that at worst you can have your IT department fix it, it should be a no-brainer.

Unfortunately, swag and kickbacks convince suits otherwise.

Not gonna happen

[Anonymous Coward]

The sad fact is while you are working the managers are out playing a few rounds of golf with
the salesmen, complete with drinks and a lap dance at the local establishment. Most companies
I have dealt with are run by IT managers that will drop a signature in a heart beat for a little
kick back.

Okay how about.

[LWATCDR]

Linux
Samba
MySQL
Postgresql
Apache
Perl
Python
Ruby
Gcc
PHP
Java
Asterisk
I think you will find all of these in large corporations. AKA "Enterprise" situations.

Re:IBM is adopting

[CannonballHead]

I've used Lotus Symphony (and use OpenOffice at home). To me, it actually seems slower than MS Office and is a little bit of a pain to work with at times. Unfortunately for me, saying MS Office was "nicer" is not a hip thing to do on Slashdot, but it's unfortunately true. At least in my case.

Useless Survey

[thethibs]

As a minor aside, the linked OpenLogic survey is useless. They only polled the people who joined their webinar--people already involved enough to be interested in a comparison of FOSS servers. That's one heck of a selection bias.

Re:Don't ask permission

[CannonballHead]

Hm. So you talk up a non-free (expensive) solution. You then watch the manager take all credit. You expect all blame to go on manager. Right. What's your credibility now? If I was your manager and you talked up this expensive proprietary product and it crashed and burned AND made me look bad, you're not going to be sticking around too long.

Re:IBM is adopting

[rmcd]

I use OpenOffice under Ubuntu (and MS Office only when I absolutely must). I agree that OO is slower and less polished. But I have found that it gets the job done, and the MS Office interface has its own issues (I'm among the hard-core ribbon-haters).

The great thing about IBM adopting symphony is that this should lead to improvements in the software. Nothing like eating your own dog food to make it taste better.

Re:IBM is adopting

[Anonymous Coward]

THe important thing to note though, is that Lotus Symphony uses ODF as its native file format. Regardless of what you think of the Lotus product, this means that the largest hardware and software company in the industry will be promoting ODF over DOC.

Re:Nobody ever got fired for...

[digitalunity]

Kind of insane really since being an MCSE doesn't mean shit if Microsoft crashes and burns and isn't around to write patches for you anymore.

At least if you went with IBM(depending on the product) there's a smidgin of hope that the community or your own developers can patch your business critical piece of software.

Not quite

[Burz]

The 'users' of a web filter are sysadmins. These expert 'users' are the ones who interface with the server and router software that runs a network.

In this discussion, we are talking about true end-users and the desire of sysadmin types to make them use a nebulous classification of software ('Linux') that only the expert can competently sort through to make a desktop work.

The management types instinctively know that what the author is trying to sell them isn't something most end-users can grasp. And that just doesn't float in an environment that normally centers around person computers and their distinct operating systems. Management might have to use this 'Linux' thing themselves, despite never really registering its Look and Feel. And they probably never will because it doesn't have one per se.

The only sure way to promote Linux-based desktops in a large corporate environment is to pitch a shift toward managed thin clients, and don't mention 'Linux' until much later. IT management understands that thin clients are a different paradigm than PCs, with the former being centrally managed by one or two sysadmins; they may even understand that Unix/Linux does thin clients well; they also won't let you anywhere near their middle- and upper-management PCs (glorified terminals are for peons).

Re:Don't ask permission

[CannonballHead]

Either way, you are basically steering the project manager into a bad decision.

If I were a manager and one of my IT guys DIDN'T warn me that this or that wouldn't work, and I paid a lot of company money for it, I'm faced with two options, in my mind...

1. My IT guy was ignorant. (not good for IT guy)
2. My IT guy (especially if I just shot down his suggestion) wasn't particularly interested in seeing a non-his-suggestion idea work.

Maybe I'd make a weird manager, I don't know, but I'd rather have my IT guy be completely honest. Either way, no manager is going to be HAPPY with their IT guys that can't get an expensive (what do they care if it's complicated, they are paying you to figure these sorts of things out) solution working, and isn't going to be happy if his idea turns out to be a bad one, and isn't going to be happy if his idea was not only not cautioned against but supported by his IT guys. Who then couldn't get it to work.

If the manager is smart, you can even allow him to be part of the solution. If not, he or she can take the fall for the expensive mistake.

It sounds to me like you are assuming a stupid manager and a genius IT guy (who, by the way, couldn't get this "ridiculously complicated and expensive" solution to work). It also sounds like the IT guy is rather arrogant... in my experience, anyways, managers tend to not like arrogant IT guys, hehe.

Anyway. Honesty seems to work. My current manager is all for doing stuff in free (and legal) ways if it actually works. And he wants me to be honest about whether or not it's going to work, how much work it's going to require me to do, how much upkeep, how many problems I foresee running into because it's free and/or unsupported, etc.

Re:Nobody ever got fired for...

[Hadlock]

Sadly this is how it actually works in Texas. Maybe not at the local level, but state education contracts are deterimned by total discount as a percentage rather than total dollars saved. Educational contractors have evolved their pricing so that their actual asking cost is 50% (or so) of the MSRP in most cases. High dollar bidding is a bizzare art/dark magic and is completely void of any reason. Fortunately I don't work in state contracts so I'm not breaking any NDAs by saying this.

Re:IBM is adopting

[cp.tar]

So open format standards are more important than overall software quality? Not sure I really agree with that.

I would.

It does seem counter-intuitive, but an open standard at least guarantees that your documents will be readable if you conclude the software you are using does not meet your needs. You simply get a new program and leave the documents be.

An open standard means a more level playing field. And that means some evolution can occur.

Re:You know--"Enterprise", "Enterprisy", ...

[jrumney]

You forgot to mention that the salesman is paying for lunch after we finish the 18th hole, have you ever seen open source that does that?. -- the Management team

Re:Premature

[mjwx]

The single biggest problem with FOSS is that there is no one to share the risks with.

No, the single biggest problem with FOSS is the illusion that MS or other proprietary vendors will share some of your risk. When this illusion is shattered, the rest of the problems are trivial.

Don't even try.

[IGnatius T Foobar]

That's about the sum of it. Big "enterprise" is steeped in the "no one ever got fired for buying [large lumbering vendor]" culture. One of the advantages of small businesses is that they're nimble and willing to experiment, especially if they can realize cost savings along the way. Bigcos only started using Linux servers after they percolated their way up from the bottom, and that's going to be the case for every new grassroots technology, whether it's open source content management, open source collaboration, etc.

Gunning straight for the enterprise is a losing proposition.