Advocacy

  Myths
  Press

Dojo (HowTo)

  General
  Hack
  Hardware
  Interface
  Software

Reference

  Standards
  People
  Forensics

Markets

  Web

Museum

  CodeNames
  Easter Eggs
  History
  Innovation
  Sightings

News

  Opinion

Other

  Martial Arts
  ITIL
  Thought


One Size Fits All...?
Standardization, good or bad?


By: John C. Welch

There is a rocketing trend in the IS world today towards 'standardization'.

First off, let's look at what is meant by that term. According to most dictionaries, a standard is a type of flag or pennant, a pedestal or base, or "a basis for comparison, criterion". Standardization is defined as conforming to a standard. However, the way the term is being used by most companies, I think "conformization" is a better term. What these companies are doing is telling all employees who use a computer, that they will all get the same toolset, regardless of posistion, unless a deviation from the 'standard' can be heavily justified, with a very good Return - On - Investment number, (ROI).

There are VERY important reasons as to why this is bad, and this article will look at them.

What comes first?

The first problem with this is that it reverses the proper order of how to solve a problem. Normally, (read, in a way that works), you first identify the problem, find the solution(s) to that problem, figure out which solution is best for you, determine how to implement that solution, gather needed resources, (tools, people, etc), and implement the solution. But this enforced standardization anagrams this procedure. Now we have the following steps: Buy the toolset according to arbitrary 'standards', ID the problem, determine the possible solutions, determine which solutions your toolset will allow you to implement, implement the solution. Of course, if your toolset won't allow you to implement that solution, you beg/borrow/steal a toolset that will. This is the "All I have is a hammer so all my problems are nails" school of thinking. The problem is, "one size fits most". How many cases have we seen, where company X starts a much-ballyhooed "migration to the open Windows standard", (Yes, I know that 'Open' and 'Windows' is a simply exquisitely ironic oxymoron, but that's for another day), only to *very* quietly later admit that NT/95 can't really do the job alone, and that they will still need their Macs, VAXes, AS/400s, OS/2 boxes etc. This is a fact of life, and it is a simple one: "Nothing Does Everything".

The fact is, you need more than just a hammer to build a house. You need backhoes and drills, and yes, hammers. Don't get me wrong, NT and 95 have their place, but they aren't even close to the all-purpose toolkits that Microsoft markets them as. Consider this: MICROSOFT doesn't even buy into this PR. Guess what they store their financial data on? VAXes and AS/400s. What do they do much of their documentation, advertising and PR with? Macs. Seems to me, if the shyster won't use the snake oil, then I shouldn't be forced to either.

Okay is not always best

The other thing is that NT and 95 are okay. At everything. NT is an OKAY file/print/application server. NT is an OKAY database server. NT is an OKAY graphics machine. 95 is an OKAY desktop operating system, (OS). If OKAY is all you can afford, or all you want, then, OKAY, go 100% Microsoft.

Me, I'm old - fashioned. I like Snap-On and Makita over Stanley. And if I am going to spend umpteen gazillion dollars to upgrade my company, then I want the best tool for the job, not settling on making the job fit my tool. If you need graphics work, get a Mac. If you want a PC-based gateway to IBM mainframes and AS/400s, use OS/2 Warp Server. If you want a good Database, use Oracle, or DB2. If you want a computer to base your company on, use a Unix server, or a Mini, or a Mainframe. In other words, USE THE RIGHT TOOL FOR THE JOB! Don't just blindly settle for "okay".

"But John, my IS staff says they can only support one environment reliably without spending a lot more money than on IS than I can afford." To quote Remo Williams: "BullDookey". If you work for a company, and the IS people use this as a justification, leave. If you can't, you have my sympathy. If you run the company, and they feed you this line, fire the lot of them, and hire some folks who know what they are doing.

My first post - military job supported eight operating systems, on six hardware platforms. In addition, the MIS department did all the telephony work, including all Moves, Adds, Changes, (MACs), switch programming, in short, anything that wasn't RPG programming, we did. When I started, I was one of three MIS techs, including the Operations Manager. When I left, the department had just ballooned. To five. We worked fourty - hour weeks, had almost no overtime, scheduled or otherwise, and we left our pagers at home on vacations! There was also a standing rule that dictated anyone who called anyone else in from a vacation would be immediately fired for incompetence.

They way we did it was simple. The most important credo was, "Make a decision, and do it". The reason this worked was because even if you made a mistake, what mattered was not the blame, but did you learn from it. Because of this, we were never reliant on only one person for anything. Unfortunately, most MIS shops don't work this way. Instead they act like Europe after World War I, and create territories and empires that are jealously and viciously guarded. As a result, the telecom people can't/won't fix the computers on their own desks, etc.

Besides, I defy anyone to show me where, with ALL related factors included, a migration to an all-Microsoft-network produced the real savings predicted. Show me the smaller IS department. Show me the great whopping amounts of space in the computer room, because you got one itty-bitty NT box to take over for that great nasty mainframe. Show me your, (now-reduced) IS staff wearing the smiles that mean they work a 40 - hour week, and can now go to all their children's school events, and have those two-week no-beeper vacations. You won't see it. As a matter of fact, if you can prove me wrong in this, I will detract one of the bad things I have said about NT over the years. I'm not terribly worried.

Who knows their own job better?

Another problem with the forced "standards" scheme is that it's, well, dumb. Hiring someone to do a job, paying them well for their skills and abilities, giving them benefits, and then telling them, "I don't care that you are the best [substitute job title here] in the world, and that you used [substitute software titles here] for years. We use Microsoft products only, and so shall you for as long as you work here" Usually, this is followed by "Gee, why do we have such a high employee turn-over?" The fact is, anyone who tells an employee how to do a job on that level is micromanaging way too much.

I do NOT advocate complete anarchy. Data must be shared, BUT, decide the format (i.e. Word8, PDF, Wordperfect8, HTML, eps, etc.), and then make sure that all your tools conform to this. Sharing data is data-format bound, NOT OS bound. I share data with 95/NT/Solaris/Linux/IRIX users all day long, no problems. What you do is spend more time planning how to share data, as opposed to doing it all ad hoc. Unfortunately, most IS departments are not given this option of long-term planning. They are told, "Do .... NOW". So, they scramble and do it, as fast as possible. Usually this does not allow itself to careful consideration of options, but rather the most obvious one. This usually means Microsoft.

So, how do you, as a Mac user, manage to live with these Windows "standards"? Well, you have to present it as a business solution.

First of all, forcing everyone to the same toolkit ONLY benefits IS in the long run. Point out that you will not be up to your current work levels on a new platform for a long time, as it took you years to get to where you are now. Point out that this is due to differences in the way the same program has to work with different operating systems. Photoshop on 95/NT CANNOT be 'exactly' like Photoshop on the Mac, due to things like color handling, fonts, (which would all have to be replaced), lack of real multiple monitor support, differences in video drivers, printing differences. If you use Quark, the loss of AppleScript means that you will have to manually do all those tedious things that are now automatic.

The important point to drive home is that in a year, you will be, (maybe) up to speed on 95/NT, BUT there are things that will forever take longer to do, and the same output will now cost more money, period. As well, that year will be forever lost as far as productivity goes, and if you have to outsource some of your work, while you adjust, then the people you outsource it to, besides costing a LOT more money, (you're still getting paid the same, and now they have to pay someone else to cover for you while you learn how to do what you were doing fine last year), will probably be using MACS!. Point out, (realistic figures work *very* well here) that your company will never get the same productivity out of you on 95/NT as they did on a Mac, due to hard differences, and that this will *forever* cost them more money, daily! If that doesn't work, then walk, if you can. I have worked for companies that won't listen to the people who make their product. It's never worth it.

Toodles y'all,          

John


Created: 07/11/98
Updated: 11/09/02


Top of page

Top of Section

Home