30th Anniversary of the (No Good) Spreadsheet

theodp writes "PC Magazine's John C. Dvorak offers his curmudgeonly take on the 30th anniversary of the spreadsheet, which Dvorak blames for elevating once lowly bean counters to the executive suite and enabling them to make some truly horrible decisions. But even if you believe that VisiCalc was the root-of-all-evil, as Dvorak claims, your geek side still has to admire it for the programming tour-de-force that it was, implemented in 32KB memory using the look-Ma-no-multiply-or-divide instruction set of the 1MHz 8-bit 6502 processor that powered the Apple II." On the brighter side, one of my favorite things about Visicalc is the widely repeated story that it was snuck into businesses on Apple machines bought under the guise of word processors, but covertly used for accounting instead.

Comments

Why use MUL/DIV

[Anonymous Coward]

When you have shifts?

What if...

[DSmith1974]

...John C. Dvorak were no longer paid to write lame articles?

Wow

[bondsbw]

Dvorak is an idiot. To use the old adage: "Guns don't kill people. People kill people."

If a bank trusts a spreadsheet based on a bad formula that is provided by the bank itself, is it the spreadsheet's fault? If the CEO chooses that saving 1 cent a year by outsourcing the call center to India, is that the spreadsheet's fault? Please.

Re:Wow

[hey!]

The problem is a phenomenon I call the "competence delta function".

A tool can make many things simple, but sooner or later you run into diminishing returns. As returns diminish, you eventually find yourself at a point where the next thing you want to do takes a step up in competence from what you have right now.

Imagine that you plot effort (on the y axis) and reward (on the x axis). If you are looking for a certain level of reward, you can read how much effort it takes off the X axis. Things like a command line interfaces have a big up front investment to learn, represented by a step function at 0 reward. However the power of command line interfaces (at least good ones), is that the benefits and effort needed thereafter are reasonably proportional over a very large range of tasks.

GUIs, on the other hand, have a low initial step, but eventually you want to do something that has to be expressed conceptually, and there you've got to eat that command line step function and then some. A sysadmin with strong shell scripting and Perl abilities will be able to achieve more than a sysadmin who can only work a GUI.

So the question for an "easy" tool is this: where does the delta function come? Really easy to use tools can be a trap if you don't see this coming. A word processor is highly useful, but at some point you may ask it to do something that would be better done in LaTex.

Spreadsheets are very easy, very useful things. Two dimensional organization of facts may be uniquely useful, giving more flexibility than one dimensional organization, but having none of the capacity for things to hide behind others that three dimensions gives. In fact, the word "plan" comes from the same root as the word "plane".

There are limitations on what spreadsheets can do of course (aside from things like: they are not databases). But I think most people run to the end of their practical, factual, mathematical or business knowledge first. For example, I can say, "prepare a budget for next year," and give you a bunch of spreadsheet templates. If you don't know what you are doing, you'll produce a really bad budget, but it will look great, just as good as a good budget in fact. The reason is, "produce a budget presentation that looks great" is on the left hand side of the spreadsheet delta, but "produce a budget that is financially sound" is well beyond what a spreadsheet application can make easy for you.

I think the same thing goes for the other bete noire of thinking managers everywhere: the PowerPoint presentation.

It's not that PowerPoint isn't a wonderful tool. It's just that it is not a replacement for communication skills. Sometime watch An Inconvenient Truth. You might disagree with Al Gore, but he worked for years on that Keynote presentation until he could really communicate with it. You could take the same presentation, and unless you were an unusually gifted commnicator, you would not achieve anything near an Oscar caliber performance.

As much as PowerPoint is not a replacement for communication skills, it is even less a replacement for critical thinking skills. During the dot com boom, I remember reading how the phrase "send me the stack" was gaining popularity in business circles, "the stack" meaning the PowerPoint slides. Some VCs were so hot to get in on the gold rush the were making investment decisions based on PowerPoint presentations rather than business plans.

A tool that makes certain things easy is always a good thing in itself. The danger is in the illusion that by relying upon it, everything will be easier. Software tools are particularly insidious, because they have the ability to make incompetent efforts look really good.

Instruction set.

[drolli]

Oh my goodness, did they really write it in assembler? I always imagined they already used high-level languages at that time.

And nevertheless, the non-availability of multiplication or division is honestly the smallest problem when programming the 6502 in assembler. Using a decent macro assembler it does not take a lot of effort to implement these two instructions. What i personally collided more with where the awkward addressing techniques of the 6502, and, of course, the quite um... limited stack, and of course, having only 3 registers sucked. I liked the Z80 much more form a low-level viewpoint. But in never though about the absence of multiplication instructions as a bad thing, just a little training....

Re:Instruction set.

[Alioth]

The 6502 wasn't that bad, or at least the 65C02. While you only have 3 registers, you do have fast zero page operations which makes it almost like having 256 registers. However, I still prefer the Z80, it makes things a lot easier to have the 16 bit register pair ops, and notwithstanding the 6502's zero page instructions, most routines on the Z80 are a bit easier to program since most of the time you don't have to shuffle things to and from RAM because you can fit everything in the two register banks. I still write Z80 asm today, it's fun.

Cannot believe I am saying this...

[DrWho520]

...Dvorak blames for elevating once lowly bean counters to the executive suite and enabling them to make some truly horrible decisions. even if you believe that VisiCalc was the root-of-all-evil, as Dvorak claims...

That which infuriates me the most about the tech sector is corporate executives building wealth upon the backs of laboring engineers. I have yet to receive an explanation as to why some VP somewhere gets to make ten times as much myself. When the company is not making record profits, it is an engineering problem. When we are raking in the dough, it is an executive success. No one ever looks to see how difficult the problem is because, they cannot fathom the problem being solved. My first day at orientation, you could tell the engineers from the financial analysts. We were in Dockers and collars and they were in three piece suits. Where did we go so wrong that support staff are the ones elevated to executive positions? Why is balancing a checkbook a more executive skill than writing the tool that tool used to balance the checkbook?!?

This only thing that disgusts me more is sharing a sentiment with Dvorak.

Re:Cannot believe I am saying this...

[Yvan256]

Not that I want to contradict your thoughts, which I share (especially the pay check part), but let's put it into perspective.

Someone might be able to write the tool to balance the checkbooks but at the same time be unable to actually make good use of the program.

Another example would be Photoshop. I'm pretty sure the people who programmed it aren't nearly as good at using it as actual artists. The programmer probably never went to design school, etc.

Yet another example would be Word, Pages or any other word-processing program. Just because you can program such a beast doesn't automatically make you an award-winning writer.

You get the idea.

Re:Cannot believe I am saying this...

[flyingfsck]

One day when you try to run your own company, you'll realize the problem with what you said above. The fact is that trying to find financing and projects to execute, in order to keep 500 engineers busy, really ain't easy.

Re:Cannot believe I am saying this...

[Hatta]

That which infuriates me the most about the tech sector is corporate executives building wealth upon the backs of laboring engineers. I have yet to receive an explanation as to why some VP somewhere gets to make ten times as much myself.

You think it's just tech companies? No sir, this is how capitalism works. Really, it's just feudalism without the hereditary aspect. You are a serf, deal with it or go get an MBA.

Re:Cannot believe I am saying this...

[rickb928]

"That which infuriates me the most about business is corporate executives building wealth upon the backs of laborers."

There. Fixed that for ya.

And yes, it's always been that way. If you want something bigger than your immediate and extended family can make, you will adopt this model. Even Communism does this, despite their protestations. Profit is the only point of a business. Humaneness, responsibility, fairness, and honesty are desireable by society, and expected, but not necessary. Without profit, a business can only fail. Even the 501(c)3 must somehow derive income, usually, and while it will show a profit of zero (mostly), it actually will probably employ people, acquire goods and services in pursuit of its goals, and thereby redistribute the income it receives. Profit is another line item non-profits either call by another name, or hide as 'reserves' until they fail or disband.

And your family will eventually resent you as well, if they don't already.

Re:Cannot believe I am saying this...

[Anonymous Coward]

Joel Spolsky once wrote:

Programmers need a Subversion repository. Getting a Subversion repository means you need a network, and a server, which has to be bought, installed, backed up, and provisioned with uninterruptible power, and that server generates a lot of heat, which means it need to be in a room with an extra air conditioner, and that air conditioner needs access to the outside of the building, which means installing an 80 pound fan unit on the wall outside the building, which makes the building owners nervous, so they need to bring their engineer around, to negotiate where the air conditioner unit will go (decision: on the outside wall, up here on the 18th floor, at the most inconvenient place possible), and the building gets their lawyers involved, because we're going to have to sign away our firstborn to be allowed to do this, and then the air conditioning installer guys show up with rigging gear that wouldn't be out of place in a Barbie play-set, which makes our construction foreman nervous, and he doesn't allow them to climb out of the 18th floor window in a Mattel harness made out of 1/2" pink plastic, I swear to God it could be Disco Barbie's belt, and somebody has to call the building agent again and see why the hell they suddenly realized, 12 weeks into a construction project, that another contract amendment is going to be needed for this goddamned air conditioner that they knew about before Christmas and they only just figured it out, and if your programmers even spend one minute thinking about this that's one minute too many.

To the software developers on your team, this all needs to be abstracted away as typing "svn commit" on the command line.

That's why you have management.

Re:What if...

[Rogerborg]

Then we'd go back to making decisions based on gut instinct, rather than what we do now: have beancounters revise their assumptions until the spreadsheet confirms our gut instinct.

bad analogy - think crank

[Reality Master 201]

Spreadsheets aren't like guns, they're like methamphetamine.

It starts out innocently enough - a couple sheets here or there - maybe a long weekend working out a household budget. It's all good fun. By the time you realize a problem, though, you're hitting the 65k row limit. You're writing VBA and macros, you're embedding external data sources - and haven't backed up your work for days. It drives you insane and causes brain damage.

Just say no to spreadsheets.

Re:bad analogy - think crank

[Richard_at_work]

Just say no to spreadsheets.

Once again the tool is blamed for the usage - there is nothing wrong with spreadsheets per se, its the user that needs to have the boundaries clearly defined.

Re:bad analogy - think crank

[Ohio Calvinist]

This is absolutely true. Spreadsheets are notorious for holding important business functions that are often designed by end users who may/may not have the best coding/design (VBA or formula) skills, that are typically never put through a QA/peer review process, and many times exist on a sole employee's desktop computer and have risen to the point that the business rules have bene forgotten over time and that sheet is necessary for some calculation for some function. These are the cases when IT gets raked over the coals when that desktop fails when responsibility lies on the user, and the user has really put the department in a precarious position. Word documents are just as bad. I've literally seen staff take a print out from the ERP system of a list of contacts who were past due for an event or account recievable take that list and manually edit the "gray boxes" in a word document that was write protected with a password that has been lost in staff turnover making it very difficult for IT to make changes when they ask for it, and then enter the same contact info in a second word document rigged to print a single label on a desktop label printer. With a Crystal or SQL report, we could have automated the process and saved time/resources for the staff and cut down our support of esoteric business processes. However, rather than work with us, one of their own rigged up something we had no idea was in practice had I not headed over there to help a technican with a connection to a server.

This is my same objection to having important business functions being run out of Access databases often developed by the most computer-able person in the department but whose skills are completely lacking. At the community college I used to work for, we had our standard ERP/student info system, but rather than approach IT to add some tracking for special programs into the system one of the student services staff started writing lame Access databases (without a single relationship mind you) to track student attendance in some program offices. What it ended up doing was causing the users to do double entry, made useful information exist outside of the insitution wide data source, and when it failed, it had become such an important part of business, IT was expected to fix a resource that was effed up from the beginning.

For the small business with 1-10 employees it is a great, in expensive way to work electronically. For anything bigger, it is trying to fish with a stick, shoestring, and bubble gum. It costs far more money to have workers working inefficiently and even worse, allowing them to stick with the skills they learned in a high school computer apps class than thinking critically, than ponying up the dough for a server or two and a more robust information system with a programmer/dept liaison to help them work more effectively. (Keeping in mind that it is possible to go into overkill mode.)

Re:bad analogy - think crank

[Beyond_GoodandEvil]

To overcome IT bottleneck (yes, it happens that IT just does not deliver)
Typically, this is reason I've seen. Of course, it is never quite that simple, more often than not, PHB wants something really fast, avg. joe tries to take initiative, does a quick one-off script etc. And no one wants to take the time to sit down and do a proper planning session for this now critical business process. Of course the high priest mentality of "only IT folks" shall mess w/ the data can lead to a different sort of bottle neck.

Re:bad analogy - think crank

[hazem]

we could have automated the process... However, rather than work with us...

I used to work in IT and now I'm a "business user" and I'm all for getting IT to simplify, automate, and improve processes.

However, what starts as a simple request to automate some simple but tedious task, or provide a way to link 2 sets of data ends up being treated like a multi-million dollar project that will need a team of 10 people 6 months and many hours of meetings (all charged to my budget) to put together a feasibility study, which will then be put into a queue where it will be evaluated next fiscal year by the prioritization committee, which, if approved (and it won't because there's a huge project consuming ALL discretionary IT resources) will still take another year and a half to design, test, QA, and deploy.

So instead of doing that, I spend a couple hours to write, test, and document some little Excel/VBA tool or some simple Access import-query-export solution that easily solves the problem, saving my employees loads of time (while typically improving their accuracy as tedious, manual tasks are often error-prone themselves) so they can focus on things like running the business.

As I said, I'd love to have IT solve these little problems because they are so often "low hanging fruit" that should take very little effort to do while saving many hours of work. But we always get told, "No", "Next Year", or "We'll need to conduct a study", when if they had just sent a low-level coder over to work for a couple hours it would have been done before I could schedule a meeting with all the "stakeholders" who would be conducting the study.

Having been in IT, I understand completely how bothersome it is to deal with things when the users go rogue - but if IT is unresponsive or provides no other alternative, then that is exactly what what they'll do.

Re:bad analogy - think crank

[RMH101]

I'm on the other side of the fence, but most of my job is in making sure that IT know what the business really want, and articulating those demands in a form that the budget holder can readily say yes or no to based on the benefit.
You're completely right, of course - this is big IT's problem in a nutshell.
What it eventually, and inevitably, leads to is a situation where department A are running some Excel/VBA/Access stuff to generate a report, and department B are doing likewise. The two reports eventually hit the CEO/CFOs desk, and the numbers don't match. Hilarity ensues.
What can you do about it? Well, if I were being glib I'd say "Don't outsource your IS function to IBM or EDS" but it's seldom that easy.
If you've got reporting requirements you could maybe stick in place a Business Objects or Microstrategy service they could use. Get your Business Analysts to help them define some standard reports, with a modicum of customisation available so they can do some more ad-hoc work. Help make sure that the canonical source data's clearly defined and in one place, not being pulled from a million different systems and hand-cranked into spreadsheets every month.
It's a big problem - I left my last position primarily to get away from big, slow IS functions and guess what? It's the same everywhere when you get past a certain critical mass of bureaucracy.

Re:bad analogy - think crank

[Geoff-with-a-G]

I wish I had mod points for you today. This is spot-on and insightful. There's so much contempt for Access/VBA/Excel solutions from the "real" IT people, who just can't believe you would want to do anything in that sloppy improper way. Meanwhile, that's the effective way to get real tasks done quickly.

I'm a network engineer, and I promise you - the DBAs and server admins who scoff at your quick and dirty solutions wouldn't be pleased if you took away their Linksys/Netgear/D-link/single-Linux-box home network solution, and told them they could either have no Internet connectivity, or "do it the right way": by getting two ISPs, carrier-independent address space advertised to those two providers via BGP, with links from the two routers to redundant internal switches and HSRP/VRRP/CARP. Might need some NIC-teaming on your desktop too, to guard against NIC or cable failure...

See how quick-and-dirty solutions seem appealing sometimes? Not because we're being seduced by the evil spreadsheet devil, but because sometimes small problems warrant small solutions.

Even his examples are wrong ...

[trolltalk.com]

"where's the evidence of improvement in the way business runs or works? Cars are shoddy, consumer goods are junk."

So Dvorak would want us to all drive the biodegradable pieces of crap cars from 1979? Those Fords and K-Cars were really awful. Then there was the AMC Pacer ... a goldfish bowl on wheels ...

Last I looked, computers were consumer goods. My laptop is a lot higher quality, and much more capable, than the Heathkit 4004 I would have had to settle for 30 years ago. Ditto my cell phone compared to ANY "portable/mobile" phone 30 years ago. And both, after adjusting for inflation, are MUCH cheaper today.

Re:Don't Follow the Link

[drinkypoo]

Not true, start tagging the story diedvorakdie or ohnoitsdvorak.

It worked once, maybe it will work again?

Anyone who still thinks Dvorak is worth reading, please go search youtube for the video of him explaining his methodology. 'nuff said.

It's not the tools, it's the morons, stupid

[0xdeadbeef]

What is actually killing the economy is the business major. There are too many people who don't know a trade going around thinking that the world owes them something.

Re:What if...

[jacksonj04]

People would stop combining it with godawful macros in an attempt to cobble together a slow and inefficient relational database with no sensible query or reporting tools and use a real RDBM instead.

Re:Loooooong time

[drinkypoo]

And not to mention the most important advance in spreadsheets in 30 years.

Graphing. CEOs can't understand numbers, they make their brains run out their ears. Having the spreadsheet program produce charts and graphs for you is the single most important advancement in accounting since language.

Aren't we forgetting something?

[Anonymous Coward]

Where's the hate for powerpoint? If you really want to blame a piece of software for spawning crappy, Dilbertesque, counter-productive executive culture, look no further.

Power corrupts.
Absolute power corrupts absolutely.
But it takes Powerpoint to really fuck things up.

30 years? Try thousands of years

[Rude Turnip]

Spreadsheets have been around for a long time; there are cuneiform tablets still around that showed how many cattle somebody had. I've got 50-year old reports in my office that have spreadsheets of financial ratios. The only difference now is that they're made on computers. Before a spreadsheet by itself can be blamed for anything, it will need to have at least as many cells as the human brain.

Unfortunately, you're a commodity

[jamrock]

Where did we go so wrong that support staff are the ones elevated to executive positions?

Forgive me for saying this, but you went "wrong" with your career choice in college. The reason why "support" staff are elevated above you and your fellow engineers (I'm assuming you're an engineer) is that they're administrative support staff, i.e. they are actually trained to run a business (or aspects of the business) and they'll be promoted within the administration of the company; whereas engineers are part of the production team, which means that engineers will probably rise only as far as project or department head. Executives build wealth on the "backs of laboring engineers" (and sales clerks, machinists, programmers, etc) because you're commodities.

I can understand your frustration, but the fact is that in any organization--large, medium, small, corporate, military, religious, political, whatever--there will be only a very few who are able to run the whole thing, and all things being equal, the qualified ones will rise to the top, provided that they're also politically savvy. An unfortunate fact of life is that there also exist within any organization the ass-kissers, toadies, and fast-talking con artists who scheme their way to positions well above their level of competence. Such glaring injustices will rankle obviously, but regrettably the vast majority of people within an organization really don't have a clue how the whole thing works. Forgive me again for saying this, but your post only reinforces this notion; you really don't know what's going on from an administrative standpoint, and I get the strong sense that you are either totally naive about, or disgusted by, organizational intrigue and politics. Good for you, if that's the case. You probably won't get a seat on the corporate jet, but you get to keep your soul.

And I'm certainly not presuming to suggest that engineers cannot run a company; my eldest brother was an engineer who worked at his chosen profession for only about a year after graduating, then went into the financial services industry and took to it like a duck to water. He is now the owner of a successful mutual fund company.

Re:Loooooong time

[KeithJM]

My friend, I wish I had mod points today. Much funnier than the parent post.

Re:What if...

[westlake]

Millions of secretaries -- I mean Admin Assistants -- would have to type department phone lists with word processors.

This is funny.

But it cuts close to the truth.

Spreadsheet planning wasn't the novelty.

The novelty was that plans could be updated instantaneously - without employing hundreds of clerics and dozens of machines to make it happen.

Re:Dvorak? Get real...

[Yvan256]

Actually Dvorak is so often wrong about Apple that you can almost be sure the opposite of what he says will become true.

Re:Don't Follow the Link

[King_TJ]

The thing with Dvorak is, one of his articles is fun to read for one of two reasons:

1. It's so patently wrong, readers enjoy putting together long replies to punch as many holes in his flawed ideas as possible.

2. It touches on some valid points, and the "challenge" is for the reader to figure out if his started results are due to reasons Dvorak outlines, or for other reasons entirely.

Re:Loooooong time

[juuri]

Graphing. CEOs can't understand numbers, they make their brains run out their ears.

Bleh. We are spatial, visual creatures by nature, graphs make complex and even simple representations of data much easier for everyone. Dunno, where exactly this whole mantra of it just being for stupid bosses came from when graphing functions were created for mathematicians.

De-facto accountants

[sjbe]

I'm an engineer. My supervisor is an engineer. Our department head is an engineer. Our vice president is an engineer. Holy smokes, even the president of the company is an engineer. The CEO? He's a bean counter.

As someone who is both an engineer AND an accountant, I can assure you that your president, VP, and even department head are all de-facto accountants as well. Accounting isn't some mysterious thing that only accountants do. If you are responsible for a budget, or handle/manage cash in any way, shape or form, you are doing accounting. Even as an engineer if you have any responsibility for the cost of the product you are producing, congratulations, you are doing cost accounting.

Accounting is simply recording and monitoring what happens to the assets and liabilities of the company. It's as integral to management as math is to engineering. You simply can't manage a business without getting your fingers into accounting. Just because your diploma says engineering doesn't change that fact that it probably is part of your job. Being good at understanding cash flows might help you get to the top faster if that is what you want but you simply will NOT get to the top or stay there without understanding accounting.

Good tool

[SCHecklerX]

unfortunately usually misused.

A spreadsheet is not a database.
A spreadsheet is not for pretty formats.
A spreadsheet should not be used for recurring analysis.

A spreadsheet *is* great for figuring out your mortgage payments.
A spreadsheet *is* great for doing college/hs laboratory analysis.
A spreadsheet *is* great for one-off, quick calc, and preliminary design work.

CYA & Real Data-driven Companies

[geoffrobinson]

So many business decisions come from CYA decisions. What you are describing is just doing some sort of analysis for CYA. "Hey, our analysis said everything was fine."

Real data-driven companies do analysis differently.

It was called VisiPlot

[earlymon]

Graphing. CEOs can't understand numbers, they make their brains run out their ears. Having the spreadsheet program produce charts and graphs for you is the single most important advancement in accounting since language.

Gee, I don't know about you, but I used VisiCalc. And its companion, VisiPlot.

I built a just-in-time inventory control system for our small manufacturing concern (about 90 parts suppliers, with lead times from 3 months to 2 weeks), tied to past sales overviews and various sales projections.

Had about 8 or 10 "standard" graphs for the boss every two weeks, showing inventory as idle, in QA, in production, in final QA, in shipping, and in repair (warranty and not).

All around 1980 thru 1982, all on an Apple ][+.

That's a good 28 years ago.

Sorry - for all of the new and fab fancy Excel features, as far as I'm concerned, they're simply not there.

The only things I've used Excel for that I didn't with the Visi series are:
1. Quick building of Fourier transform tables when I was just too lazy and hungover to code them up
2. To increase my vocabulary of cursewords (OK - that's not possible, I'm lying, I'm from Detroit) trying to get simple x-y plots without markers

And to the idea that the new spreadsheets provide statistical functions: big whoopie deal. Sum_y, sum_y_squared - you're done, and you've used VisiCalc. (Then again, I'm a snob who believes that you can't work the teleography of statistical equations you've no right to be spewing about statistics anyways.)