Slashdot Mirror
Do Software Versions Really Matter?
An anonymous reader writes "I work for a rather large software company and I am currently working on a completely new product. So new in fact, that the official name has not even been decided. I had assumed that the version number for this product would be 1.0 (at most). However recently I learned that the Product Managers want to release this NEW product with a version number somewhere between 5.0 and 8.0 because 'there is a stigma about buying 1.0 products. People assume it's no good.' This latest Dilbert-esque comedy routine nearly sent me over the edge. So to gauge my sanity against that of the upper Product Management, I ask the community: Do version numbers play a role in software decisions, or have product version numbers lost all credibility and meaning? Would the community feel comfortable buying version '6.3' software (and paying tens of thousands of dollars for it) knowing that it was the first release of the product?"
Let me know when you hit 7.0
Personally, I take to opposite view. If I try an application labeled something like version 6.0, for example, and it still has a lot of bugs in it then I'm likely to be a lot more pessimistic about the software. After all, version 6 software ought to have most of the bugs worked out by then. I would think poor quality at version 6 would reflect much more negatively on a company than at version 1.
We've all been conditioned by a source that will go unnamed for now that version 1 software is probably full of bugs, so it's not unexpected. It's also probably true that some people will avoid software simply because it's version 1. Yet, it's the same software whether you call it version 1 or 6, so it has the same bugs in it (e.g. the user who tries the software will experience the same problems, regardless of the version label). For a company to risk losing the good will of the customer on a marketing gimmick seems foolhardy to me. Trust is easy to lose, hard to regain.
The NSA: The only part of the US government that actually listens.
Turn it up to 11!
That will inspire confidence in quality...
Most users won't even notice the version number unless you put it in the face. Just call it FooBuster and put the version number in an about box somewhere.
A lot of us are probably using Open Source software that's been released and relatively stable for years but is still only at version 0.2.07 or somesuch. We're not exactly representative of the general public.
See: http://news.cnet.com/8301-10784_3-9814858-7.html
When Oracle began selling its first commercial SQL relational database management system in 1978, which version was first officially released?
A: Version 1.0
B: Version 2.0
C: Version 3.0
Answer: Version 2.0. There was never a 1.0 version. Said Ellison: "Who'd buy a version 1.0 from four guys in California?"
This is like the one where they had to rename the movie "The Madness of King George."
Americans, the story goes, wouldn't be interested in "The Madness of King George III" because they missed parts I and II.
Way back in 1995, I upgraded my version of Windows to Win v95 from Win v3.11. I thought "oh man, there's been 92 upgraded versions of this software! I better get with the times!"
The Internet is generally stupid
If there's a version 6.3 of software in my field that I've never heard of, I generally assume it's some crappy shareware knockoff of what I'm already using.
If it's version 1.0, I want to see what was so important that they had to make a new piece of software (which is why I tried out Google Chrome).
I have seen purchasing decisions based on version... but usually it has less to do with what the version number is, and more to do with how long the version has been on the market. If a version 1.0 was just launched, unless there was a large business case for taking the risk of buying it, the company I work for would wait until 1.1 (or until 1.0 had been on market long enough to prove itself stable). Same goes for upgrades, a new release of a product is not moved to unless there is a large business case for the move (or the version has been on the market long enough).
What's long enough? Depends on the vendor and their release cycle.
I mentioned tinker-toys once in a post - now I'm modded down for life.
Over the past five years, Version 9.6 became 9.7 with no real updates. 9.7 jumped to 10, and then 10.2 with no real updates. Then it jumped all the way to 15.7 with no real update. Then came version 16.0, with no real updates. Next month I can look forward to version 16.2! I'm not expecting any real updates.
Your employer basically just admitted to you that they're trying to deceive and mislead the customer.
The reason people feel more comfortable with higher version numbers is that they assume the code is more mature at version 5 than the first cut would be at version 1. Anyone with a serious interest who heavily depends on the software will see past this and look into the history of the software, especially where large amounts of money are changing hands to aquire the software. Your company on the other hand is hunting for schmucks who'll give them money without doing proper research. Not a good sign. That is not how you gain long term customers and cement a relationship that will result in further sales and on-selling. Your sales/marketting people probably already have their CV ready. So should you.
These posts express my own personal views, not those of my employer
There never was a Dbase I version, their initial release was Dbase II. :)
Who pays attention to version numbers on anything nowadays? I don't they've all because ridiculously named and hard to keep track of, ME, XP, 2K, MX, CS1/2/3/S, Gusty Gibbon, Feisty Fawn, Hoary Hedgehog etc etc.
What happened to the good old days when it really was just simple version numbers?
Can I leave this box empty?
would be to take the 4.5.1.2 version and rebuild it from scratch to a new 1.0 version where all the old cruft is removed.
Windows 7 is actually the .1 release of the third version of NT. (No wonder they finally gave up and just called the next version "Windows".) But then they started the NT line with the first release being "3.1".
Going back in history, dBase II was actually the first version of dBase. For just this reason: no-one trusts a 1.0.
In open source, it goes the other way - the project has to just about take over the goddamn world before they'll admit it could possibly be a "1.0" release.
Summary: version numbers are marketing just like everything else.
http://rocknerd.co.uk
Maybe that is the reason they didn't name it an Xbox 2, when there is a Playstation 3 out.
God spoke to me.
What you're proposing simply won't work, and carries a huge risk of making you and your company look dumb. Also, without a plausible explanation why your 1.0 is actually labeled 6.3, the customers, sales force, and techs are all likely to make up their own. Many of them are not very appealing:
A) We actually stole it from a competitor and kept their version numbers
B) We went through six major version changes before arriving at a marketable product
C) We have been selling this product to a different market, under another name, for years
The '1.0' moniker is a label. It carries with it the meaning that something is new. Remove that label, replacing it with one that means something is NOT new, and people's minds will invent the reason why.
Unless of course you come up with a good story and get it straight ahead of time. This is well known as a basic tenant of dishonesty...
- Release it as a beta, and never let it out (Charge for the "beta.")
- Use the year as the version
- Use a chemical element or gemstone as the version
- Use an animal as the version.
- Use two random consonants.
- Periodically drop the most significant digit
You have a lot to learn about business. As a Software Engineer, your best approach would be to make software products that your company can sell. That means you listen to sales and marketing, and anyone else who knows what its like trying to shift copy on the ground. When they say you have to release as version 6.3, that's what you do. If 1.0 doesn't sell, you're out of a job.
If I see a version-greater-than-one of something, I'll take a look at the change list and see how quickly new features get added, or bugs get resolved.
So if I see a high version number and no history, I see a scam.
I don't go for scams. I prefer to report them to the local authorities.
I remember long ago when Slackware jumped from 4.0 to 7.0, not because there had been 3 major revisions that just hadn't gotten released or something like that, but because Red Hat was already on 6.0 and Patrick Volkerding was tired of being asked why Slackware wasn't at 6.0 yet.
To answer the original question, version numbers don't mean much. They can give you an initial clue, but you've got to look at the history of the software to know the truth. Sometimes there are huge version jumps just because, sometimes there are major changes but only a change to a minor revision number.
I concur with leaving in the About box, and nowhere else. The only people concerned with looking for a version number would be those who need to know if this is an upgrade from their current version.
The name of a product is a marketing decision, period. The version numbers that make sense to you as developer of the product, at best, mean nothing to the buyers of the product. At worst, well, your own example about "1.0" is perfect.
You need to have some internal scheme for keeping track of builds and versions of your product for release management and support issues, but there's no sense in having engineers decide whether a given release is 2.5 or 3.0. Let marketing pick the name that's most meaningful to buyers.
Are you adequate?
GAAH! MY PRINTER IS ON FIRE!!! PUT IT OUT! PUT IT OUT!
I'm going to assume you're an Engineer. (Since you're a Slashdotter and refer to "the Product Managers".)
I think it's swell that you're all involved with your project and everything. That said, do you like it when management and/or marketing types get all in your shit about how you do your job?
Honestly, those cheese-eating motherfuckers probably really do have a better idea than you do about how to sell this stuff. Let them. You'll all feel better if you do!
-Peter
Why use software X version 1.0 when I could use software Y version 6.1?
Some people just see the bigger number.
And those people are called PHBs (or CEOs), and sadly they sign the checks and often make the decisions. Of course by the same token if given the choice between software X 1.0 by company Z which just took the CEO out to an all expenses paid lunch at some fancy restaurant and gave him a shiny handout claiming the software would do everything he ever dreamed of, and software Y 5.4 by company T that their IT staff says is the better application but the CEO never heard of, odds are their getting software X 1.0.
Curiosity was framed, Ignorance killed the cat.
You missed a perfect opportunity for "Post 1.0!!!".
Interesting coincidence, yesterday I was reading that Adobe Premiere Elements 7, released this month, is the successor to Premiere Elements 4. It seems Adobe wanted the version number to be the same as Photoshop Elements so that it wouldn't be deemed inferior, also because they're trying harder to get people to buy the two softwares in one purchase.
Personally I think it's pretty ridiculous. Integral version numbers are supposed to be indicative of development milestones, not to rate the product. However, the higher a version number, the higher the chance to be a well-established software. I think this is what they're getting at. They're basically lying on the age of the product to get more respect, like an underage boy with a false ID card hoping to get in a strip club.
That's cheating and should be dealt with accordingly: "Adobe, go to your room! No more Elements for 3 years!"
I'm a little bit older, so to me the suffux "2,000" sounds all futuristic and spacey, despite the year 2,000 happened 8 years ago.
Because I'm older, I probably would be in the position to purchase software, so such gimmicks would probably work for me.
Of course, if you really want to get my attention, you need to add a prefix to the software title. Let's say you're developing software that integrates certain proprietary database formats into a universal format, and you call it "Spectraview" or something.
Spectraview 1.0 sounds kind of low rent.
Spectraview 2000 sounds shiny, as if it was developed with NASA technology.
But add a prefix, like somebody's name, and you get something like "Tom Clancy's Spectraview 2000". Which sounds pretty darn cool.
You can tell your marketing guys this. Claim the idea as your own. I don't care. I'm still waiting for "Jane Austin's MS Paint 2000" to be bundled with the next version of windows.
The Internet is generally stupid
A product family is used as a generic name for the whole family and if you don't have version numbers then you have hell.
Customer: I'm using WizzoProg and getting this problem.
Tech support: Which version of WizzoProg are you using?
Customer: WizzoProg. I couldn't find a version.
Tech support: Ok that must be the original Wizzoprog.
Five minutes of confusion....
Customer: Oh, you remember you asked for the version, I can see know it is V3.2.
Engineering is the art of compromise.
This isn't really quite as dilberty as the poster indicates. This is a symptom of a more general problem, which is that non-OSS software almost always sucks, because the economics dictate that it has to suck.
First of all, let me throw one big [citation needed] on there. Your whole argument starts with the fact closed source sucks because it's closed source, and open source doesn't because it isn't. You hint at something about economics, but that looks like hand waving to me. With an opening argument like this, the rest of your post is surely going to be just as fun...
If it was OSS, users could install it on their machines, try it out for a while, and decide whether it was any good or not.
Closed software does not automatically mean you cannot try before buying. Quite a few closed source applications have free trials or even free versions. And there are surely Open Source products that are not free. For example non-commercial clauses. I think you are making the mistake of confusing Open Source with free. Which actually makes you a fairly uneducated OSS Zealot to boot.
(Note that this still works fine for commercial OSS. E.g., people can try Ubuntu before deciding whether to deploy it widely in their organization and then pay Canonical for support.)
Again, you are mixing up OSS with free. One could imagine Windows having a free trial. You should think about if yer really talking about OSS or cost.
If it's not OSS, you don't typically have any way of knowing whether it's good or not. Sure, you could read reviews, talk to friends, etc. But that's sort of like deciding to buy a car without having a chance to test-drive it, just based on your buddy saying he has one and he likes it.
So wait a minute. More hand waving here. How exactly do you know if software is good or not by the virtue of it being OSS? There's the tired of argument of 'Well you can read the source code!' Yeah right. How did you decide FireFox was good? Did you read all the source code? And even if you are crazy enough to do that, who else is? No, you probably heard about it word of mouth, just like you would with closed source software. I think again what you meant is, 'If it's FREE you can try it without paying for it.' However, see above.
The worst piece of non-OSS software I ever owned was Adobe PageMaker 6.5, but the only way I found out how bad it was was by writing a book using it, and finding out after I'd gotten pretty far into the project that PageMaker was gradually starting to corrupt my files, and was also crashing often enough to cause me real problems. It would crash one day, and I'd lose my file. So then I'd open the file again to page 93, which I'd been working on, and it would crash again because page 93 was corrupted. So then I'd get the file back off of backup. But then I'd click to page 87, and it would crash again. So the backup was no use either, because it was corrupted on page 87. In this example, there's absolutely no way I could have tested the software sufficiently before buying it to find out that I was going to have these horrendous problems.
So how long should a free trial be? I think what you want, again.. is free software. You -never- want to pay for it. Maybe you'll make a donation later after you've used it a few years. Maybe. And as for the long sob story about losing your data, if it's closed or open source, could have the same bugs, and still lose your book. I don't see how this is, again, any sort of argument for OSS.
Because users usually can't evaluate the quality of non-OSS software very effectively, there is absolutely no incentive for non-OSS software houses to work on quality.
Wow. I'll have to remember that. As long as I keep my code closed, I can write crap and people will buy it. Oh
I was going to ask the O.P. the following questions. How does a salesperson respond when a prospective client asks:
1) "What are the new features in this version as compared to the previous version?"
2) Or, "We want to compare the new release to the previous release. How can we get a copy of the previous release?"
3) Or, "We'd like to contact current users of the package. Can your company provide a list of current customers whom we can contact?"
4) Or, "Please provide a list of all of the service packs and patches released for the previous version, the time from when the problem was identified to when the update was made available and whether the update resolved the issue."
I could go on but I think everyone sees a pattern here. Making the first release of a product version 5.0 or some such nonsense works as well as most lies. The only way to maintain the lie is to tell more lies which then beget a need for still more lies. Eventually, it all unravels although current management may be under the impression that they can take the money and run before they're found out.
Cheers,
Dave
They that can give up essential liberty to obtain a little temporary safety deserve neither safety nor liberty.
Ben
If it starts with 6.0 and I happen to know it is a new product I begin to doubt anything else you claim about the product. I expect those I do business with to display a high degree of integrity and this displays the opposite. Customers do not like to be lied to.
I maintain Finnix, a system maintenance livecd. The first release was 0.03. The next release was 86.0. Why?
1) Why not?
2) See 1.
3) It had been 5 years between releases.
Finnix is currently at 92.0, and I've got to make a decision about version numbering soon. The reason is simple: "There Will Be No Finnix 95", for obvious reasons. I may just jump from 94 to 100.
I've noticed that, when Finnix is on a X.0 release, people tend to transpose it incorrectly a lot more often, saying "Finnix 0.92" etc. I think many people just cannot comprehend a version number greater than 10 or so.
If it's version 6.0 and it's buggy, you've just shot yourself in the foot and no one will want to buy future releases of your product. If it's v1.0, then there's more leeway given. Version hopping is for those who have learned enough to be dangerous. Those who don't know are blissfully unaware of the version of their product. Those who do have the knowledge don't upgrade just because there's a new version. It's those in the middle that equate newer with better - and they quickly learn their lesson as their wallet is emptied. This is a prime example of a business that's more interested in immediate profits than it is in retaining a loyal customer base. I'd suggest looking for employment elsewhere.
What would YOU do when you tried to research FooBar v6.0 ... and could not find anything at all about v5 ... v4 ... v3 ... v2 ... v1 ?
My first thought would not be that Marketing had fucked with the version numbers. It would be that that company's past product have sucked so badly that NO ONE would use them.
If I cannot find a SINGLE user who is happy with v5 what does that tell me about the likelihood that v6 will be decent?
And when I find out that v6 is really v1 ... but Marketing wants to fuck with the numbering to FOOL people into buying it ... no way. I'll go with a competitor's product. That's just too many warning signs for me.
Much depends on the customer. Most are quite stupid. But then you say the price the over $10K. Then I might expect the customer will at least have asked around. They will quickly find out that even with the version at 6.3 there ARE NO OTHER CUSTOMERS. They will then think "something is wrong" suspect fraud and bail.
If I knew it was a 1.0 release then that would explain the lack of existing customers but a 6.0 with no existing user base? I'd be thinking "scam".
But on the other hand if the software were to be sold retail in a box as an "impulse buy" for $29.95 thaen you could expect you customers would fall for the scam and maybe never even find out. But with a $20K price they will at least try Google to find reviews and the like and when they come up missing... not good you've just lost the trust of a potential customer.
I totally agree with the parents post.
Another option for the original poster would be to name their product after the year, as in
FooBar 2009, or even more vague as in just plain FooBar
..........FULL STOP.
There's actually 2 versions of a product. One is the marketing version, this can be anything FooBar 2008, FooBar 8.04 (representing months) FooBar 2008 sp1 patch2 etc.
The other version number is really required for support. This needs to be able to specifically identify the build/patches applied to be able to provide the customer with help when they run into trouble. It can even be a build number. It's not sprayed all over the product as that is what the marketing version number is for. But maybe there's a particular screen or command line utility that provides this information.
The beauty is that if you recognise both, then you don't even need to enter into the marketing debate. It's nice to be able to say, "You can call it whatever you want, we just need to know x months in advance so we can put it in all the right places before testing".
If you want a specific example of this, look at Internet Explorer. It's marketing version is Internet Explorer 5 or 6 or whatever, but if you go to Help/About you will see a version number 5.02.0123773 which is probably a build number or something.
hth
ws
So does Anonymous Coward have good karma?
I was reading about the history of Lockheed (spelled Lougheed back then). The Lougheed brothers named there first aircraft Model G for similar reasons.
Having said that, I'm afraid that Upper Management will soon be engaged in a massive rebranding exercise as they find that the market shuns a certain software product from a particular company that displays "1.0 quality" when the version number reads 5.x or something. They will probably have to completely rename and rebrand the product and perhaps even see damage to the name of the company as a whole.
I'm sorry to say that that it looks as if that's the way things are going to be. You see, rebranding products is something that Management understands, and they might just be happy to look forward to an exercise that falls squarely within their core competency. I don't think that damage to the name of the company will impress them however, since that typically is a long-term thing. Longer than their likely tenure with the company in question anyway.
Having said this, it's not unlikely that you will see an urgent demand for bug-fixes (apart from the usual demand for additional features) as the product meets headwinds in the market. There is a chance that this will enhance your job security, highly desirable in an economic downturn, so don't look down your nose at it.
What you might do is start thinking about what type of defects the product is still likely to exhibit (despite your best efforts during development, testing and debugging), what additional features are most likely to be demanded, and start thinking about how to go about fixing those bugs and implementing those new features plus how much time / effort that would take. Then when the defects emerge you can impress your boss with a calm but supportive attitude, and well-thought through plans that offer him alternatives and allow him to offer sensible options to higher management.
Besides which, it's not unheard of to run a book among your co-developers with bets on what general type of errors will be found and what priority Upper Management will attach to them. Only, be sure not to let Upper Management catch on, as they will then insist on placing bets themselves and will adapt their priorities in a way that will make their own bets come true. Be warned!
A few years ago, I worked for a network operations center at a university, and we managed the internet access of over one hundred thousand users (mostly the university interconnects and the internet gateways, not everything down to the dorm room or anything like that). We were toying with the idea of using a ticketing system to handle issues that cropped up, and I was asked to evaluate some open source software packages.
Eventually, I found Request Tracker, slapped together a demo server, and showed it to the "Director of Technology." He stroked his beard. "It's okay," he said, frowning, "but the ticket you just created has the ID number of 1."
I shrugged. "Well... yeah," I said. "It's the first ticket."
He shook his head. "That's not going to work. We need to be able to start it much higher. Otherwise everyone is going to know that the software is new."
I stared at him. "We get phone calls from about two dozen network engineers," I said. "We're on a first name basis with all of them. I think the giveaway will be that they get a ticket number at all, not that it's low."
But he was adamant. I was annoyed enough by the whole conversation that I stopped working on it, and for all I know they're still not using a formal ticketing system. (Which is probably just as well, because even if they'd started a ticketing system at id # 0, four years later they'd probably be into the low three digits.)
This is nothing new.
Remember dBASE I? Neither do I. dBASE II was the first one you could buy.
Mal-2
How is the Riemann zeta function like Trump rallies? Both have an endless number of trivial zeros.
BEA tried this awhile ago with their brand spankin' new 8.0 version of their server. The marketing department got clever and instead of calling it 8.0, decided it would be released as version 8.1
It might have initially fooled a few people, but it later just served as fodder to ridicule the company when we'd uncover the most basic bugs.
More importantly, their marketing department must have eventually concluded it was a bad idea, as the next version abandoned that scheme and was released as 9.0 (http://en.wikipedia.org/wiki/Weblogic)
Obviously people will want to claim that they are not influenced by silly and trivial things while operating in a business capacity. But the sad fact is that people do tend to buy for irrational reasons and they also do not buy for irrational reasons. There are all kinds of ethnic and almost spiritual locks between people. When I worked in sales I often new I had a big order as soon as hellos were exchanged with a potential buyer. Often sales have nothing to do with the price or quality of a product. To stay alive companies need to use whatever edge they stumble upon that is not immoral or illegal to push sales.
While version 1.0 does have a stigmata 5.0, 7.0 or 13.0 also have stigmas of their own. Market research has shown that software with lots of numbers tend to scare off customers also. Most customers will shy away from them because they feel you'll be coming out with a new version soon, and drop support for the one they want to buy. This is why several company's have dropped revamped their numbering system (eg Adobe photoshop CS instead of 9.0). I suggest not advertising the version number of your software which is fairly common, and will make to look like less of a fool if your potential buyers actually research before they buy.
Here you're ignoring the argument of my post, which is that you can tell whether it's good becauese you can install it for free, and test it as much as you like.
If that's your argument, then it's really weak. Why didn't you even bother to address the fact that much closed-source software (probably most) come with trials, demos, or free versions?
You argument appears to be based on a lie. Yes, I can install and test proprietary software before I commit to buy. If the software is crap, I won't buy it. Don't you think that potential sales are a good incentive for companies to write non-crap proprietary software?
... and then they built the supercollider.
Compromise and ask marketing to call it "Captian Placeholder 2009". This way nobody lies to anybody but can still give the impression that your product is modern and fresh while providing versioning in case of later releases. Somebody seriously considering giving you so much money for software will have found out that there where no prior releases anyway and if that doesn't turn them away they probably ask themselves why you feel the need to lie to them right from the get go or how in the hell you need 6 releases to have something you are willing to actually sell.
... X] like SAP does with their netweaver family. Their first app server was "SAP Netweaver [snip] 620" and 620 can mean anything they like here.
Another route to go is to use version numbers that are not obviously in sequence [1.0
___
No power in the 'verse can stop me
If you have a strictly linear update model, you can get away with just about any numbering scheme you want. This is a rare situation. Usually you'll end up maintaining multiple releases, such that version 6.0 is not a strict superset of version 5.4 in terms of features and bug fixes. If your customers are smart, they'll see through any numbering games, so you should just be honest with them. If your customers are stupid, make sure you get your bonuses in cash, because there's no telling if the gravy train will last long enough for your options to be worth anything, because you're completely at the whim of the marketing gods.
It may be worthwhile to call your first beta version 1.0, and work up to a 2.0 commercial release, to appease some PHBs, but if your customers distrust you so much that management is seriously considering calling the initial release 8.0, lying to them so transparently will only make the problem worse.
Rather than making shit up like this, just call it Widget Tycoon 2009, and use build IDs for patch levels. If all goes well, you might even be able to get your customers to pay to upgrade to what would otherwise be point releases every year or two. Who wants to be running Widget Tycoon 2009 in the year 2011?
There's no failure quite as dissatisfying as a complete and total solution to the wrong problem.
I'm always astonished with say open source software that has been around for 10 years and is still at version 0.62 or something. I like the premise of version numbers = major changes, subversion numbers = minor changes + those features that were planned for the version but didn't make the release date and sub sub version numbers for patches. I hate when things are released and the version number is essentially just the build number without any hint of how major the changes were.
I would check the software history first.
If the software history is short (in your example, it does not exist at all!) so the updates does not bring lots of fixes (if there is lots of bugs) or new features (if such are needed), I would not buy such software.
If I would really need your software, and there is no competitor, I could then buy it... mayby!
But lying for customers even in VERSION numbering... it is very stupid. I would say to your marketing people to grow up and stop lying for customers, because they are the "thing" what keeps you up and going. Play nice, be honest and respect the truth.
Even that no one cant say that how much you need to update your software, until you can change one of the X.y.z numbers, but it would be very stupid if just one "normal" feature would grow up version X+1.
I just dont trust corporations if they try to cheat me by using marketing in wrong way. It is very dirty trick and I take it as offence against me and deal is off.
These sheninigans have been pulled by software companies for over a decade. Frankly, I doubt anyone puts any credence in a version number anymore -- as the whole process is corrupt.
It used to be that 1.0 was the first version (or perhaps, even a sub 1). Now, 3.0 sounds low, so, we start at 6 or 7 (nice numbers, right?).
This disrespect to convention, has made the whole version numbering irrelevant. On top of that, some companies don't use version numbers but use names, letters, or other such monikiers.
So, who cares? Call your product Foosoft blue.
> have product version numbers lost
> all credibility and meaning?
to me, commercial software companies have lost all credibility and meaning, and that's the real problem.
I already expect that they will not deliver carefully engineered products, but throw the first thing that compiles on the market, and then just see what happens (that's v1.0). Many products are broken by design and are implemented carelessly, so they have a lot of bugs. But actually, if version 1.0 isn't good, then in all probability version 5.0 will not be much better - because you can't fix a broken design with workarounds. Version 5.0 may even be worse than 1.0 - because if the implementation is really bad, noone knows what's really going on in that code - so you may fix one error and introduce two new, more complex errors with your workaround.
On the other hand, if you can proove, that your product has a good design and is implemented correctly, I don't care about the version number - I'd even use a version 0.32 in a production environment.
Pick a sensible version numbering scheme and stick with it.
My personal favorite is x.y.z, where x is incremented at major rewrites and when incompatible changes are introduced, y is incremented when new features are added, but backward compatibility is preserved, and z is incremented for maintenance releases that don't add features and don't break compatibility (except when caused by buggy behavior, of course).
This way, you can tell that your configuration file for version 1.1.0 will work on version 1.1.1 and version 1.2.0, but not necessarily on 2.0.1, and that if someone is running 1.1.0 when 1.1.1 is already out, they may be running into bugs that have been fixed since.
Whether you, as a user, trust a given version number to represent a reliable piece of software is, of course, entirely subjective. It is wise to remember, though, that there is software with 0.6 version numbers that is rock solid, and software with 9.0 version numbers that is junk. More or less the only thing that can reliably be deducted from version numbers is that higher numbers indicate newer versions of the software. A version number doesn't say anything about how good the product is.
Please correct me if I got my facts wrong.
Who's going to be making the purchase decision on this turkey? Power users? Clueless end users? Developers? MBAs?
Generally speaking, the more technical a person's mindset is, the less stock he'll put in the version numbers.
If you're selling to developers and high-end power users, you can call it version 0.1, and they'll probably just think you're cool. But selling to technical types is a double-edged sword, because you've got to give them technical information about what it does and why its features and capabilities are superior to other products that do the same thing, or they won't be interested. Also, if they do buy it, they will keep wanting you to make the product more customizable, so they can reconfigure it to meet their needs better.
If you're selling to people with an MBA and no technical knowledge, on the other hand, the version number *does* matter. It's not the most important consideration, but it matters.
Cut that out, or I will ship you to Norilsk in a box.
Why not call your product "software name 2009" You can use that naming convention for a long time until your product reaches maturity. Don't make available anywhere the version number and put out a press release talking about how long the company has been in business and how much expertise they have in the area that this software is designed for.