Slashdot Mirror
Business Software Needs A Revolution
An anonymous reader writes "According to a Businessweek Online article, today's high-end business software is bloated, buggy, and too expensive - no surprise to those of us who have paid our bills by adding pointless features to some piece of software arbitrarily priced at $100k. Evidently, firms are now re-evaluating their software purchases, and finding that they're not working out the way the sales guys told them they would."
I'd love to see a survey of how many people use the huge number of convoluted and complex review and version features provided by Microsoft Word. The addition of these feature seems to represent the only major change from one version to the next of this microsoft suite, nowadays...
One of the key problems is that software vendors think that they should continue to add more and more features. Each time a software vendor solves some little bullshit problem for one customer, they decide to throw it into the next version resulting is feature creep. This might be kind of cool for the geeks but it sucks for most users, especially the typical users of the software. As most of us know, as you increase the number of features, you increase the complexity. As you increase the complexity, you decrease the usabilty. Thus, paradoxically, as you help some people you hurt a lot of other people. Stated another way, the harder these vendors try to help users the more they hurt them. Usability just keeps dropping.
How to Download YouTube Videos
Evidently, firms are now re-evaluating their software purchases, and finding that they're not working out the way the sales guys told them they would.
re-evaluation doesn't mean refund. It means lets spend more money and hope it works this time! (or maybe the existing vendor has an upgrade solution! we can use our relationship to get a great discount!)
The revolution has to come from the businesses who buy the software. And the sound of that revolution would be "I know what I want, I want a,b and c". Venders would be in utter shock, many will fail because they are not used to actually making what is demanded only what they think/know it should be.
To be honest, at least 50-90% of the cost of big software packages goes into maintaining another company, paying that company's CEOs and sales staff, paying for first level support people to misdirect your call and other things that are, to a great degree, unrelated to the quality of the software you're getting.
Think about it: for $100k, you can get package X, which does half of what you need it to do in some areas and twice what you want it to do in others. Or, you can hire me & my buddy Josh for a year. We'll write you a custom piece of software integrating open source tools, work right along with your employees and give you all the code and a support contract for XxX hours over the next YyY years.
If there's an OSS package that already does most of what you need, you can probably hire their developers to customize it for you quickly and at a very minimum expense. You don't even have to tell anybody about your custom code, unless you intend to release the binaries outside your company.
And of course, if you can get three companies that need a similar piece of software, you can invest in a small business that does exactly what you want and split the cost. That's how my friend's firm works...the bills are paid for by the big guys, and anything they sell on top of it is a bonus. As a result, their rates are 1/2 to 1/10th those of their pay-for-our-big-name-CEO competitors.
That's your software revolution: customization, adaptation and competent small businessmen. And it's already happening.
Hey freaks: now you're ju
The big business app vendors have mastered the buzzwords to impress the CxO's and boardrooms into believing they solve the problems. After all, if FORTUNE 100 software company whose software is used by everyone doesn't fix the problem, who can? The evaluators are shut off because the sale is predestined by the owner.
I saw this happen when my company evaluated a $2 million package by Big Software Company X and went away saying no way in hell. Then it came down from above to look into it, and $4 million and 2 years later it's still not done. The problem is, any project with that much money, and the big names on it can't (by definition) fail. So more money, more time, more frustration.
Of course, it's easy for someone so close to the implementor level to see it as management's fault. They turn around and see it as the implementors' fault for not doing it properly, since it works everywhere else so well.
They overruled the mechanics and bought the Jaguar, and don't want to look foolish and admit to the neighbors it's always in the shop. Articles like this are a positive sign though...
The article suggests that all business software should be sold as a service. Doesn't this give remote access to the most vital part of a business' infrastructure? Is it possible to serve software remotely without creating a huge security risk?
I've got this kind of situation at work. The department I work in is replacing the system I've written and am now maintaining, with a different system that's less suitable for the job and costs 4 (yes, FOUR) times my salary, per year.
Isn't bureaucracy great...
Does anyone believe that C#/Java will improve the overall quality, standardization and total cost of ownership of enterprise software? Or is the improvement only incremental? My theory has two parts: 1. Strongly typed, managed code environments will eliminate many of the problems created under older C++/VB/COBOL programs. 2. The scope of the runtime environment, which now includes XML "web-services" and other high-level constructs will help to eliminate integration problems that called for customization in the past. I've always felt that enterprise software vendors could make their biggest improvement in sales by reducing the overall implementation costs and leaving the core features alone.
I work at a pretty large .com, one who actually survived the bust and maintains a profit, and has a pretty significant amount of traffic. We have used ATG Dynamo for our application server for several years, partially based upon the built-in ability it has to do an MVC architecture, personalization, pools, and so forth.
However, we just completed a web application that was built using many open source components, including Struts, Validator, JUnit, and others. By using open source components we have completely divorced ourselves from using the proprietary technologies used by Dynamo, and have opened ourselves up to the possibility of using a different, and of course cheaper, application server. This would not have been possible were it not for stable, performant open soruce initiatives.
Not only is management happy because we have (potentially) saved a bunch of money, but the developers are happy because they are much more friendly towards open source than closed technologies; it is far easier to get an answer to a question via Google than it is to pay for and go through the hassle of using a support contract of some kind.
I do not mean to denegrate Dynamo at all, because it is actually a fairly good application server. The licensing costs, however, just cannot be justified when so much of the functionality provided by it is already available elsewhere, for free.
I'm going to take this somewhere the original poster might not have intended, but it's something I believe. The correlation between speed and stability is the ability to use higher level langauges. The more powerful machine, the higher the level of the programming language can be. There's a limit of usefulness, perhaps - but we haven't gotten there yet. The more the language can do to take care of buggy software, the closer we are to bug-free software. Common programmers will NEVER write bug-free code with the languages we have. Where we can get is to the point where bugs are strictly logic errors, and not semantic or syntax errors.
for salesforce.com courtesy of BusinessWeek.
Don't read into what BusinessWeek "predicts" too much. They are almost always wrong. As an example, they were cheering on Enron's business plan as being brilliant and unstoppable when they should have been questioning their numbers and trying to uncover their pyramid scheme. As I recall, BusinessWeek had pictures of the Enron execs jumping on trampolines and laughing in the article - business journalism at its finest!
I think traditional software still has some legs yet.
Exactly what Jakarta subprojects would you suggest for building an ERP system? Or CRM? Compiere? Please.
:)
I just spent 7 months with Epicor (ERP and CRM), which is one amazingly crappy piece of software. But where's the open source choice? I mean there's not even a viable OSS replacement for Quicken let alone a ERP, CRM, or real accounting system.
If you want tons of consulting bucks, write a *good* open source ERP or CRM platform and sell the consulting/support/training. But until there's a decent *enterprise* choice, we're stuck with the crap from the vendors.
One IT manager told me "All ERP solutions suck. And whichever one you choose sucks worst."
sPh
Ihhaa...
Luckilly comes open source to the rescue with mature buissness applications that just don't mimic the old crap but comes with new ideas.. oh.. Never mind..
Show me a CRM that is not a crappy pile of MySQL and php version Alpha Alpha Beta, or so... Why don't people abstract the database away so you can connect MySQL or ORACLE. It is not like these applications stores more than VARCHAR and INTS.. or whatever.. Metabase seems to get things straight, but hey that is just php..
.. but to make a real interactive program you first have to pick Gnome, KDE or maybe Kylix or whatever.. gaahh.. it is not like you can ask any of the maillists which to pick without some serious asbestos fatigue..KDE or Gnome, which is best ?
.. did anyone say RAD.. didn't think so..
It is actually a chicken-and-egg problem to see who is at fault, the buyers or the vendors.
Let's start with the buyers, and their proxies, the press. What is the easy way to evaluate software to purchase? Surely not to take the time understand its architecture, algorithms, efficiency of code, etc. and to fully test whether it works in the environment for the intended purpose. No, it is to compare features. So, we got the "Feature List from Hell" and the trade press replete with exhaustive feature comparisons. As if they meant something.
So, as a vendor, to what are you going to build and sell? Of course, you want to be the first to be out there with just enough of each and every feature have the most filled column in the reviews (it barely matters that the features actually work). Sure, you'll also make noise about the rest, but we know it is all Marketicture. E.g., Microsoft has been talking about the benefits of code reuse since the 1980s and implying that Office was more efficient because of common elements, yet StarOffice is about the only suite that actually implements an OO component model.
The article makes an example of Oracle's bad release of 11a, because it was rushed to market. They overlook that this is repeating history -- Oracle almost went out of business in the early 90s because their software was so fundamentally rotten that they had major lawsuits from both customers and shareholders. Obviously, even the industry leaders don't learn. Or maybe they do -- Oracle did it to gain market share, and it worked; at the end of that period, the competitors with better products were fatally wounded, but Oracle "fixed it all" and came back. The lesson is obviously: "it pays screw your customers".
Then, after years of vendors rushing to market and bloating their products with useless bells and whistles that one in 10K people might ever use, and IT managers buying it all uncritically, we then get a new phenomenon.
Consolidation happens, and a few vendors gain market hegemony. Some exploit this by starting to create deliberate incompatibilities. Now, the purchasing decisions get taken out of the IT managers' hands by the business managers. Their primary concern is compatibility -- "I tried to send a critical file over to Bob Jones at our biggest customer and he couldn't open it -- I've had enough of this import/export nonsense, so, damnit, we're standardizing on Microsoft Office for everyone!". In the mid 90s, the major sales forces sold directly to top management; the goal was to go around IT, and it worked.
By not being critical and business-focused in the first place, IT management lost what little power it had. They had become plumbers. Then, they got outsourced to India.
Add to that a collection of bad or self-serving decisions on industry standards, and the mess is compounded. We now have TrueType fonts used broadly, but the more sophisticated Adobe fonts used by the serious graphics experts because John Warnock would not agree to Bill Gate's demand for zero royalties on the fonts shipped with Windows. Or, did you ever have to contend with all the incompatible International character sets and code pages on a variety of browsers and email clients? Everyone talks a good game of international standards, but when it comes right down to it, there is no one standard that is actually used everywhere -- local code is still needed for every locale. And, there are dozens of examples like this.
And, of course, this is all happening in an environment where there the vendors bear no responsibility for their product working. Marketed under "licenses" that would make a pirate blush, they can peddle crap that would generate FTC prosecution for fraud if it wasn't laughed off the shelves first. Do you know anybody that wants any kind of serious device, like a car or a plane, running a PC O
It amazes me when I work with software 'solutions' that have cost millions of dollars that have no interface in or out of them other than the specific stuff provided by the vendor. I'd kill for direct access to the underlying DB or a nice clear way of moving data in and out, or a great way to make custom GUI... but the company is more concerned with ensuring that we are locked in FOREVER than with providing the tools we could use to make their software more friendly to our over all IT enviroment.
One of the reasons developers don't always use the components available from Microsoft, Sun Microsystems, and IBM is because they fear vendor lockin themselves. Of course, most custom apps do depend on compontents from all or some of those vendors, locking you into all of them and lending lie this aspect of the article in a not so heartening way.
Not that your point on the vendor API is invalid, generally you only get a halfway useful API if you specify it in the contract. Though you might get lucky if they used a clean CORBA API (or equivalent ORB) internally. Something like Maya's MEL might be nicer though, you wouldn't even have to write the shell bindings.
You won't find a good OSS timesheet system. You won't find a good OSS ERP system. OTOH, there's a whole lot of good OSS "infrastructure" out there, like JBoss, Apache, PostgreSQL etc.
The problems lie with the specialized applications. They might be built upon OSS tools (a lot of the apps I work with/on are), but the system itself is proprietary, complex (models a specific business process), has rather few users compared to other apps (like MS Word, Mozilla etc), and offer no "hobbyist value" (you won't be installing it at home, even if you had the software, since it wouldn't be useful to you at all).
Going OSS would only allow your competitors to get your work for free. Consider a system for collecting timesheets. You have to be able to integrate with the various other systems the HR dept have, maybe some card/punching terminals etc etc. The competition is FIERCE, and giving away your solution to the competition would be a commercial suicide.
Add to that, there's a whole lot of legacy systems, third party libraries, tools etc. you simply cannot give away. One application I worked on recently still use a OS/2 compatibility lib on WinNT, and the app would be useless without it. Making it OSS would make no sense
"The" business application is a do-it-all monster, like MS Office; and since it must not only do everything but do everything INSIDE everything (like actively linking spreadsheets inside word processing documents, and insane tasks like that), trimming it down is an enormous task.
Maybe the answer is to re-examine software inter-operation and see if these things can be moved further apart. I forget which open-source app was moving towards XMLizing all the data formats, but that's a step in the right direction. Next, in my opinion, will be XMLizing user output formats.
The end result? Application A, instead of having its code intermixed with application B, instead sends the embedded information to it, receives bitmap or vector data as an output, and places it in the user's view. Yes, it's a hell of a lot of added overhead; but the amount added would be worth it, in order to have nice, discrete, stable applications.
The market is demanding new bells and whistles. How many people would purchase a new version of MS Office, if it would look exactly the same as the previous one, and didn't add any new features? I've been to Microsoft sales presentations. All the people there, like me, were there to make a decision for their company to purchase the latest & greatest to come out of Redmond. You should have heard the Ooohs and Aaahs as each new (and completely pointless) feature was presented.
Well, from your description, it doesn't sound like a groundswell of demand for new features, it sounds like a bunch of people being sucked in by the usual MS marketing hype. We use MS Office at work, and most of the people wouldn't upgrade to Word 2K even though it was available -- '97 was good enough. Eventually the dept. AA got miffed because the reports coming in didn't work with her Word 2K, and the dept. manager made everyone upgrade. Lack of backwards compatibility is why most people upgrade, not some craving for new misfeatures.
You may have said "500" as hyperbole, but when we were trying to help on one Fortune 500 company (which shall remain nameless, and I can't name their product because you'll know who they are...) replace a legacy ERP system, I was shocked to find that the database record for each customer had over 300 data fields, slowly accumulated over a period of 30 years. Some of them no one knew what they were for anymore, but the majority actually had some obscure purpose in some dusty corner of their bloated bureaucracy. Each department predicted dire consequences if they lost their data fields, so the majority ended up being retained, with several hundred thousand lines of new code written on top of the modern system to effectively emulate their old one. It cost $millions to do this. This company was hurting badly because their profits were being eaten up by bureaucratic overhead as their market share migrated to the cheaper products of lean, efficient Asian companies. They're still hurting today, and I don't think they'll be Fortune 500 much longer.
I'm the parent post you responded to, and I want to respond: actually, yes, one of the key features of the 'service model' internet based software that I develope is the guarantee that my clients can, at any time, through several different very well document interfaces (SOAP, html tables, CVS, custom SQL read-only queries) pull their data out of my application.
I felt it was a gamble, and it has been. Very public API allow competition to easy write routines to suck your data out and steal customers. But the benefits and customer satisfaction have easily outwayed any negative aspects AND loss of revenue.
Granted, I'm in a niche market, with a small number (7) of very serious competitors. Maybe I'm lucky. But I do know this: already, my open approach has allowed people to integrate other software packages and 'services' with mine, allowing them to save time and money. They pay attention to this, and due to the open nature of our relationship, they share this info with me, and I recommend it to other clients. I've even formed relationships with other services that my clients use, and we make sure we can work together. I've found that due to the openness of my process, not only is my service a 'service' it is also a data deposit, because people have become confident that they can move data about as they please, and I have made it easy for them to do so.
It amazes me when I work with software 'solutions' that have cost millions of dollars that have no interface in or out of them other than the specific stuff provided by the vendor. I'd kill for direct access to the underlying DB or a nice clear way of moving data in and out, or a great way to make custom GUI... but the company is more concerned with ensuring that we are locked in FOREVER than with providing the tools we could use to make their software more friendly to our over all IT enviroment.
:-)
I have a secret... there's this little known movement in the IT and IS industry called Open Source... and it solves all this.
Ok, sarcasim aside. If you are not a broken record to your PHB's preaching Open Source, Open Source, Open Source then it will never happen. that damned sales tool based on filemaker 5.0?? yes you know that piece of crap they you pay through the nose for, the company could develop a web based replacement that uses MySQL for less. Yes you heard me. for less. Hiring 1 perl jockey for 1 year at $42,000.00 is 8,000.00 less than the software support agreement with that piece of crap that always is broken or losing data...
until companies get managers that actually have management abilities, Executives that are actually interested in creating growth and increased profitability instead of running arould doing damage control we will be stuck with the absolute worst software on this planet for completely insane prices...
I work in the media sales field... and ALL of the software from the nielsen and scarborough data tools to the sales management software to the traffic and billing software, it is all utter crap written by horrible amateurs that are NOT programmers in Foxbase, filemaker or visual basic.
and our company happily pays for this junk year after year after year.... because nothing else is available, and I cant write it as the company will simply steal my software from me. (and no, I am not going to do it for the good of the company until they pay me for working at home (I.E. double my salary...))
vertical market software is always extremely low quality. and in some markets it's worse...
On the bright side.... ther eis no way in hell they can get rid of my position in the company because of it.
Do not look at laser with remaining good eye.
Some years ago, I heard a story about a Siebel Systems account that ended up costing $100 million before it stalled because nobody had any more money to budget for training the end users after the crap was installed.
My immediate reaction was:
1) If Siebel is this incompetent, he should be out of business.
2) If his client was that stupid, they should be out of business.
3) Who authorized $100 million for a software project without budgeting for training end users?
4) I should go to the client and tell them, "Hey! For a fixed fee of $10 million to me personally, I'll solve ALL your software problems!"
It also reminds me of the story in (IIRC ComputerWorld years ago) that Travelers Insurance was suing the Cobol standards people for coming out with a new standard when the company had yet to finish converting from the OLD standard to the LAST standard. This was a company that had a TWO HUNDRED FIFTY MAN COMPUTER SCIENCE DEPARTMENT WITHIN their IT organization and they couldn't convert their millions and millions of lines of COBOL from one COBOL standard to another in less than something like ten years... My immediate reaction was, "Who is the moron running their IT department - and why isn't it me?"
Management morons...'nuff said...
Richard Steven Hack - This sig is TOO GODDAMN SHORT TO DO ANYTHING USEFUL WITH! MORONS!
I can collabrate this. We had a client that _insisted_ we add a feature to our commodity trading product because it could save them TONS of money!
They wanted the system to not allow trades that were unprofitable.
When confronted with the choice between accurate data or data that looked nice and would never ever reconsile, we were asked to have the software calculate the future prices of the commodities so they could have both...
"Sir, no offense, but if we could do that we'd never have sold our software and we'd all have retired by now"
Their own IT staff understood our frustration and it took them and our consultants a couple weeks to explain to the users why they were requesting something that was impossible.