Slashdot Mirror
The Death of Licensed Enterprise Software?
tfsm writes "Andy Singleton wrote a short, interesting article about the looming death of traditionally licensed, proprietary, enterprise software over at The IT Manager's Journal. In it, he talks about the declining revenues of software giants such as Siebel. There are several causes, but one, he suggests, is erosion from Open Source offerings."
Open source is taking a dent, but the biggest threat to enterprise level and vertical markets is the products themselves.
Typically they are low quality with bad quirks. So the company finally get's sick of it and has the replacement software written in house to replace them instead.
Now the company OWNS the app they rely on and pay's less in "annual licensing fees" for the maintaince of the software and codebase by having on-staff programmers that are NOT dedicated programmer positions.
I.E. the IT/Programmer is very common today. you change printer toner, install a new PC and add a feature to that Billing application, or squash a bug in the shipping application.
Corperations are now demanding IT and IS people that are capable of all aspects and are expected to perform all aspects. That "programmer" is expected to be at the office at 3AM to deal with a crashed database server.
It's silly to pay $60-100K for a programmer that you have to try to keep busy when you can hire someone that has good programming skills, good IT skills and actually understands Electronics at the board level for the $50-$90K. and usually get an employee that will happily work his butt off because of the diversity of the job.
This is my observation from work here. All new hires in the IT department MUST have some programming skills in C, Java, python and PHP. We intentionally do not hire anyone that has been a "programmer" or "developer" except those that have experience in OSS as a developer. But they also must show a proficiency at IT skills and prefer that they have some EE background.
this has lead to over 15 enterprise apps being replaced with in-house versions that work better and are far FAR cheaper in the long run even when ignoring the fact that it is an asset now because the company owns it instead of a liability when you "lease" or "rent"(buy) software.
The great part is that versioning systems like Subversion integrate so well with linux,OSX and windows that it takes less than 2 hours to teach a new recruit how to use our system and get them up to speed in checking in and out code.
Enterprise apps are starting to become in-house customized projects, and THAT is the biggest threat to that "business model"
Do not look at laser with remaining good eye.
My skepticism grows with each time I see this topic posted somewhere. The deathknell of the proprietary has been hailed as a Roman general marching onward to victory, but let us remember this: despite the rosy forecast, proprietary enterprise software packages remain.
While OSS has inherent advantages, the non-savvy end-users of business systems prefer lack of change to robustness of operation. If they didn't, OSS would be more prominent on the enterprise level. We tried to implement various builds of *nix and X window system on workstations at my place of work, and there was REVOLT IN THE AISLES.
The business world is not yet ready for the intellectual (not to mention monetary) cost of full transition to OSS. As much as I love it, the end-user isn't ready. It should remain the goal of all OSS developers to give these people more and more reason to change their minds.
Let us keep in mind: there are other reasons for the drop in revenues for these proprietary software vendors as well.
The Crimson Dragon
It just costs too much. At some level, people are comfortable running a database software just because it is Oracle or DB2. This is not terribly different than the "nobody gets fired for buying IBM" mantra back in the day. However, if you get just one enterprising geek in your outfit this *can* start to fall apart. Surely, there are applications where the support that a company like Oracle can provide can be the difference between life and death but, just as often, one of the open source DB's can fill the role.
At my old company, we were using Siebel on Oracle that ran on a big fat Sun system. We were still in "startup" mode and spent over $300k on that bullshit. The CEO had a grand "vision" for taking the company to a new level and we had to build our IT infrastructure aggressively to support it. Well, a year later we had a E350 with 4 procs that spent most of its time idle and 20 people working in the call center. I had argued that what we needed to "plan" aggressively instead of "spending" aggresively and had been laughed outta the room for being "short-sighted".
A year later it was satisfying to go to their bankruptcy auction....
There's no free ride anymore, it's time to innovate or die.
Depends on which side of the coin you are on. If you are on the consumer side, choose F/OSS and it can be a free ride. Unfortunately, shrink wrap software companies probably are going to have a hard time paying salaries of programmers so if you program, you'd better start liking jobs where all you do is tweek F/OSS for "customization" for your site.
So let's see you hire some high IQ people and start thinking up new ideas and industrial progress will be off and running again after a short stall!
And if you're a shrink wrap house, you'll pay these high IQ salaries with... what exactly? If you *do* come up with something great, you'll have 100 SourceForge copycats within a month and they will erode your market.
F/OSS is the great poison pill of software. If anyone comes out with something that is good (and it isn't you), then just put some effort into a F/OSS "alternative" and poison the whole market... basically make it where if *I* can't make any money in that market, then no one will.
Simply, the enterprise software vendors themselves. At this point, they'd all have to be wheelchair bound given how much they shoot themselves in the foot.
Over the years their prices have risen out of sync with target client business revenue, activity, and need not to mention the changing economic scene.
Their software often seems written explicitly to confound the most experienced users and administrators and effectively prevent any ease in enterprise-wide roll-out, installation, upgrading, and administration.
Their licensing models bear no relationship whatsoever to the realities of the usage of the target businesses, and frequently are outright hostile to newer technology usage such as multiprocessor workstations and thin clients.
For instance, I have yet to have a single installation of any Computer Associates offering go smoothly, or anything that might be mistaken for semi-smoothly. A demo copy of Unicenter once hosed a workstation I tried it on. Given the models, methods, and practices currently standard in the world of Windows programming, and the prodigious resources of CA, that takes Herculean effort to do.
Remedy ARS anyone? I'm sure this can't be the only software with an interface that would make a sadomasochistic OS/2 2.1 adherent's blood curdle.
Siebel? I worked for a company that tried their code. We lost 500% productivity almost overnight. Everyone rebelled by continuing to use Remedy ARS. You have to write some horrendously bad stuff to make people prefer RARS to your offering.
Open Source is of course, NOT a solution. Any corporation that isn't run by some weird eccentrics is going to avoid paying a code cowboy team to customize apps of all kinds, in all places in the business, and then pay their legal people overtime to make sure they are in compliance with three or six different open source-ish licensing models. As it is, there are major corporations shovelling massive greenbacks into Redmond to get Windows source access to get custom builds for their desktops. Or were when 95 was the standard. Now they might just put up with the comparatively less quirky WinXP Pro and pay a few junior desktop nerds a whole lot less.
Many companies today, trying to cut costs everywhere, are removing a lot of very useful software that their people got very comfortable with and were very proficient with, further eroding productivity. How sad is it that the vendor of the gui has overpriced it to the point that their client would rather do without and simply make use of the command prompt interface of the routers and switches instead?
All in all, things are not in the same way they used to be seven years ago. That does not mean however that Open Source is going to be the magic solution. OS still costs money. Programmers and support personnel and trainers do not work for free. I think neglect of taking that into account is the single biggest blindspot of the OSS boosters and if they don't stop acting as though the fruits of others' labors should be free on a silver platter and come with no cost, they will blow a golden opportunity to expand the usage of software in big business and simply hand it back to Microsoft, Siebel, etc.
If my grammar and spelling are off, I am [distracted/tired/careless] (take your pick)
You put into words what I'm thinking.
I have a love hate relationship with F/OSS.
I worked for a company with a proprietary software that had problems competing with *free*.
I love fiddling with F/OSS as a hobby and I smile smugly when Linux makes Microsoft squirm.
What hypocrasy on my part.
The one advantage that I can think of is that F/OSS counteracts outsourcing to India.
The work that moves to India is large proprietary software projects.
Customization of free software packages stays in the West.
The bill to the customer is probably comparable in many cases. The software may be free, but the customization costs more.
Please feel free to visit the open source project that I have in my sig.
The Internet is full. Go Away!!!
Sure, but look at the price you pay. We didn't want to renew our support contract with Oracle, but they said we had to because other departments in our company had support contracts with them. Apparently they fear that a company may have only one contract and redirect problems from all databases to the one with support.
We also have an Ingres database running in an HP-UX server. The server had a failure and the CPU was replaced. The CA support rep heard of this and they sent us a bill for $45000, discounted to $15000 if we paid without fuss. They charge by CPU power and the replacement we got was more powerful than the old one, which had been discontinued, so we had to buy the software again, even if the old one ran perfectly in the new CPU.
Enterprise software not only is expensive, but they'll try to milk you dry.
Right, open source is hurting companies like siebel because they don't know how to remain competitive in the new market. If anything, open source software is creating entirely new vertical market spaces everywhere you look.
For example, I work with a college career services office and, 3-4 years ago, all of the college office specific online systems were handled by centralized inflexible giants like eRecruiting and Career Connections. They offered very little if any direct access to data. No one was all that happy with them, but it just would have cost too much to develop internally hosted alternatives.
Now other companies, like Symplicity, have entered the marketspace and are licensing software that the offices themselves can run. They use php, pear, mysql, and their code is riddled with other lesser known free software like HtmlArea. Because of all of this open source software, development and maintenance costs came down to the point where they were able to tap into a previously non existent market of office hosted career services software.
Seems to me there is a link between F/OSS and popularity of the problem space, especially among programmers. I've never seen it discussed before, so I thought I'd bring it up.
To create a solution for somebody for nothing, I would guess you would need a bunch of qualified people to write and test the code. Since these people are not getting paid, then it would have to be something that these people are interested in solving.
Since these people by definition are programmers, they're going to be interested in stuff that programmers are interested in. So the evolution of F/OSS will continue along the lines of stuff programmers like -- encryption, database, file sharing, photo editing tools, etc.
It's going to be awful hard to get groundswell support for some new system to categorize ear wax, for instance. You can make the argument that so much of software is just the guts and not the business logic, but that's the whole point of software abstraction to begin with, so it's a non-starter.
So to me the question is: who's going to care enough about mundane, boring, business-rules based code to keep it up to date? Certainly not me -- not for free. And therein lies the limits of F/OSS.
The people who directed the reaming, of course, have long since moved on to other employment, won by their 'stellar' results.
"That's what contract negotiations - and ultimately lawyers - are for."
It's calling "screwing the customer after they've committed to your product".
City College of San Francisco is replacing their Integrated Library System with another vendor's. They explicitly wrote into the contract that the new system had to be able to access the SCT Banner college MIS system. This is not a huge technical problem, but software companies don't like having to modify their product.
The vendor proceeded to send them trainers and the like preparatory to migrating to the new system while doing nothing about the interoperability.
Now the vendor says "it wasn't in the contract" to be interoperable - now that the library is within a couple months of their previous vendor's license being up.
I could replace their ILS in 6-12 months with an OSS version for less than one-quarter of the price they paid for this screwing. And then I'd make a bundle customizing, training users on, and installing it for every other college library in the country.
But the college will probably knuckle under to this reaming. Primarily because the college has already knuckled under to SCT who charges them $150K a year for "support", but who have to pay a consultantcy ANOTHER $195K a year for ACTUAL support. This consultantcy also gets to recommend themselves for re-contracting every year. Nice racket.
This is how the software industry works.
And according to the article referenced by the headline article, corporate management is getting tired of it.
Richard Steven Hack - This sig is TOO GODDAMN SHORT TO DO ANYTHING USEFUL WITH! MORONS!
Siebel is an example of realy bad software with really good marketing. If you use it straight out of the box it is reasonable. If you customize it to meet you business process needs, you are in for a world of hurt. The company I work for spent over 20 million last year keeping alive and to continue development. My job is to keep it alive and running along with several other enterprise apps. Siebel support is the typical vendor game of send me a log, did you check the configs, turn the log level up, can you send me a log, someone will get right back with you. Repeat above sentence until your own techs solve the problem. Scalabilty is a joke. It scales just add hardware and lots of it. Our average workstation footprint for the thin client runs almost a gig of memory. Server side the thing is a pig per user for memory. The AIX port of the product actually has a windows directory and a windows registery emulator. If you run one of these kind of products the best money you can spend is in to hire a top notch DBA.
Siebel spends most of its user week telling the consultants what to do. All of the keynote speeches say this was hard, expensive and difficult but it was really worth it. I COTS software was supposed to cure those things.