Monday, January 7, 2008

Does Open Source Have Commercial Viability?

On my usual search for opinion on Online Accounting, I came across this local post by Bevan (here). He is impressed with Xero, however he is critical with their use of Microsoft .Net. He believes (perhaps somewhat biasedly) that it would have been easier to build using the open source tool "Drupal". This got me thinking: "Does Open Source Have Commercial Viability?"


I think not.


Here's why:


A business needs assurance of continuity. Most business owners have no real understanding of "IT". It is a mystery to them. And so like other mysteries, they look to follow the norm, which (today) is Microsoft. Microsoft Windows and Microsoft Office. They are the benchmark (albeit a low one at times), but everyone compares themselves firstly to Microsoft. So a business owner will be wanting to stay with what they believe will give them continuity, which is Microsoft.


A business wants simplicity. Microsoft have carefully and successfully managed to be the single provider of the greatest portion of a businesses IT needs. Once a business owner first strikes incompatability issues, they quickly realise that standardisation is the best panacea, and since all of their computers come with Windows & Office, and all of their staff are experienced with that, and their IT people are Microsoft certified, it is just so much simpler to use what is already familiar to them.


A business has money. This is unlike "consumers", who often have unlimited amounts of spare time to spend on their home computer or MySpace or personal website, but no money. A business is prepared to pay. In fact, paying for a product or service gives comfort to a business owner, as it is what they expect as part of running a business. Unfortunately, "Open Source" carries the stigma of "it must be free!" And so a business owner will subsequently treat it as anything else presented as free - it doesn't have any value.


Over the years, I have worked in all aspects of IT, from selling, installing and training software and hardware, to developing desktop and web based solutions, and more recently as a Global Product Manager. There has only been one time when I have been convinced that Open Source had a place in a business, and that was in the emergence of the internet in the late 1990's, when there simply was no other commercially viable and usable alternative.


Having said all this, I'm quite open to hear your opinion.


Stuart

42 comments:

Bevan said...

Hi Stuart. Thanks for pinging me, I let others know about your blog post.

dgtlmoon said...

yeh i mean, take gnucash for example, great opensource application, but still has dozens of highly critical bugs, some todo with basic things like unable to set the font sizes in invoices!

and if there was a bug in gnucash, where do you go?

and firefox? sure IE has dozens of issues but i bet it prints correctly!

i think drupal would be a poor selection for an accounting package, and ive been involved in drupal development for 1.5years!

Jim said...

Does Open Source have Commercial Viability? Oh yes -- much more so for the consumers of software than for the producers, though.

In the particular case of Xero's service, the use of Open Source that Bevan is suggesting is basically functional; it would not be visible to the consumers of their service per se, but if Open Source is truly "better" they would end up presenting less bugs to their users, and would be faster in producing new services, **for the same developer investment**. But that's not really what you seem to be talking about ...

What do you think Commercial Viability is? Why do you think that desiring it leads one away from Open Source?

I'd like to take a few of your comments (hopefully not out of context) and challenge them a little :-)

> A business needs assurance of continuity.

Continuity of what? And for how long?

"Commercial" software only exists as a supported service for as long as the producer remains interested (i.e. until there is a "mandatory upgrade" or perhaps they go out of business completely). After this point no real support of the software itself (i.e. bug fixing or functional extensions) is possible at all, either practically or even legally.

"Open Source" software exists as a "supportable" service in perpetuity, from a legal aspect. From a practical aspect, Open Source and Free (that's "capital F" Free, not free-from-cost) software has a significantly longer proven shelf-life than anything else.

> So a business owner will be wanting
> to stay with what they believe will give them continuity, which is Microsoft.

I agree with you :-) However, blind faith is not a laudable attribute in any context.

> Once a business owner first strikes
> incompatability issues, they quickly realise that standardisation is the best panacea

Yes -- standardisation on published Open standards for their data storage, but not what is aggressively sold to them, which is standardisation on whatever Microsoft want to sell today. Tried opening a Word version 2 .doc file with any software legally available from Microsoft today? It won't happen.

> it is just so much simpler to use what is already familiar to them.

How much does the desktop environment change between each version of Windows and indeed of Office itself? The idea of "persistent familiarity" is only really promulgated by Apple (and that not even perfectly)

> A business has money.

And therefore should spend it? No; if a business has money it should be returning it to the business owners -- shareholder dividends or re-investment are the accepted methods.

> Unfortunately, "Open Source" carries the stigma of "it must be
> free!" And so a business owner will subsequently treat it as anything else presented as free - it
> doesn't have any value.

Who said that Open Source has to be free? I'll happily charge you for my time if I produce something for you; the fact that I'll license it to you under an Open Source license doesn't change that. I'll even contract to never release the code to anyone else under any license if that's what you're willing to pay for -- but you still have the product under an Open Source license, so you have complete legal freedom to be free from *me* if you want.

And that's what I see as the most Commercially Viable aspect of Open Source software; you can become Free from your supplier.

Jim said...

dgtlmoon: "if there was a bug in gnucash, where do you go?"

Anywhere you like -- you have the legal freedom to engage any programmer in the world to try to fix the bugs.

If there was a bug in MYOB, where do you go? MYOB, or nowhere. Hope they stay in business longer than you do!

Anonymous said...

I posted a showstopper in MYOB RetailHospitality around 2 years ago. To the best of my knowledge it has not yet been fixed.

Do Open Source programmers have a committment to quality software? Collectively, yes. The thos is one of merit where errors, inefficient or downright bad code are not tolerated by the community. "Bad" programming practises and errors are gradually distilled out due to the pressure of many eyes.

Do Closed Source programmers have the same commitment to quality software. Not universally. The ethos is one of getting the next release out or moving up the career ladder. Once you get locked onto a release path where you have to prioritize bug fixes and releases, you lose the committment to ongoing quality.

Does OpenSource have commercial viability? I could not operate my business without Open Source code; I could not operate my business with Microsoft products.

goldboy said...

> And so like other mysteries, they look to follow the norm,
> which (today) is Microsoft.

Interesting. Looking at the businesses that are on the same floor as me the norm would be Linux... MS comes last along with Mac OSX. Although the talk suggests OSX will over take MS shortly.

dgtlmoon said...
> yeh i mean, take gnucash for example, great opensource
> application, but still has dozens of highly critical bugs, some todo
> with basic things like unable to set the font sizes in invoices!

You mean you can see the list of bugs? Man... what will they think of next?

Next they'll be publishing the code and the bug fixes as they are applied to that code.

http://svn.gnucash.org/trac/timeline

But it looks like they already have. Looks like the last 16 patches from 4 different developers were applied a few days ago.

I've not checked, but is there a public bug list for MYOB or Xero? Is the code publicly available so that if users have the ability they can submit patches to fix what's not quite working?

> and if there was a bug in gnucash, where do you go?

http://wiki.gnucash.org/wiki/Bugzilla

You'd start there.

Laura said...

Continuity? I was using a major professional video editing platform that was EOL'd by the vendor. Companies go under all the time. And when that happens, you are stuck with no future in your investment. Not so with open source.

Simplicity? Microsoft? Seriously? There's no question many open source products can benefit from better usability, but that comes with time. Microsoft has been around for how long and still I can't get Word to format a simple outline without going all crazy on me.

Money? Money counts only if the vendor spends it on the product ... and spends it effectively. Money isn't only measured in dollars, though. It's also measured in time. In the end, not many corporations can keep up with an active open source community. Why is Linux outpacing Windows? Why can't Internet Explorer even now render valid web pages properly?

Pointing at accounting as the representative sample is perhaps a poor choice for drawing blanket conclusions about open source. There's a whole world out there where open source is the winner or the strongest comer in the market.

Funny all this, posting this discussion on Blogger. I bet Google was used by at least one person in this thread. What platform do they run on? Or is Google not an enterprise-level commercial. Anyone viewing this on Firefox? Any Linux users out there? Hello world!

Stuart Bale said...

To anonymous (... I posted a showstopper in MYOB RetailHospitality around 2 years ago. To the best of my knowledge it has not yet been fixed ...):
If you could possibly provide more information to me, I am more than happy to investigate and provide feedback. Feel free to email me directly if you prefer privacy at stuart.bale@myob.com.

Stuart Bale said...

Laura
Thank you for your comments. I concede that very often there are faults in Microsoft products - like you, I have had to deal with countless numbers through the years!
As you suggest, my thoughts are indeed based around my experience, just as yours are. I have worked with hundreds of small businesses in New Zealand in all industry types. I have worked with hundreds of Accounting Practices worldwide. I have worked with hundreds of software developers around the world. Hence, my opinion is reflected on what they tell me, on the problems they have, and the solutions they have chosen.
And so, (in my opinion) until the 'Open Source Community' can uniformally present answers to the points I have made, then the general market won't be interested.

Of course, having said this, I think we are on the brink of a major revolution, with Internet connectivity and adoption rapidly increasing, that will see a significant change in the way people use computerised systems. But I'll save those thoughts for another blog entry.

David said...

I'm intrigued by your position on Open Source software, Stuart - mostly because it differs so totally from my perspective.

I started an open source software business back in 1998. In fact, we were your neighbours (you're at MYOB, right?) for the first couple of years until you moved out of town. We provide Linux-based business solutions (managed file/email server infrastructure, VPNs, web hosting, etc.) and Drupal web application development and hosting services. We charge real money... and we make real money. Business is booming despite the fact that we don't feel the need to use any Microsoft software (although we make open source software work with it, when our customers request it).

Our customers quite like the fact that they can rely on us and the open source products we provide for continuity. Our open source systems - based on discussions with customers who have come from Microsoft-based infrastructure - are
a) more cost effective (TCO of 1/2 - 1/10th the cost per annum what they previously paid for MS Windows licenses + Commercial Support),
b) far more reliable,
c) more secure,
d) more flexible,
e) and less hassle (i.e. simpler)
than their Microsoft counterparts.

Since 2002, our company has used the open source LedgerSMB accounting system (full client-server with a proper SQL database backend), which Bevan refers to as "SMB Ledger" in his blog entry you reference (it used to be called "SQL Ledger"). It works very well. It's a rather impressive product which offers many features not offered by consumer/small business-focused proprietary packages like Quickbooks, MYOB, and others as well as local enterprise level packages like Accredo, et al.

Admittedly, LedgerSMB is not as well documented as some of these packages, which can be a trap for non-professional accounts people. But, I pity the enterprise unwise enough to employ an accounts person who doesn't understand accounting and finance... From the support side, we offer hosted versions of it for Christchurch-based businesses. We're quite familiar with it now. We've already got a handful of companies using it quite happily (their accountants seem to like it as well).

Regarding your point that Drupal is not a good choice for building an application, and that Microsoft's .Net framework, I suspect each person backs the horse he knows. I know I'd never build anything important with .Net. Other than its many shortcomings, it's owned by Microsoft who will only ever use it maximise their profits through lock-in and avoiding true open standards. They are an unscrupulous corporation, and I personally would never do business with them if I could possibly avoid it.

Your points for doubting Drupal (and open source in general)'s ability to provide a viable framework for a Xero-like application are familar but largely incorrect. They seem to have a lot in common with Microsoft's anti-open source marketing guff...

Here's the simple point. I agree that Xero and other "Web 2.0" and proprietary desktop application developers are likely to make money for a while (like 12-18 months). But they should remember that it only takes one open source developer in the world to decide to take them on (often, however, they work in teams, like the tens of thousands working on Drupal, to amplify their reach and effect) by developing an open source equivalent application.

Typically, the open source one is faster, more secure, and doesn't have a lot of annoying activation codes and per-seat licenses. It will tend to use open standards where they exist. It has absolutely no incentive for lock-in.

It might not have as good documentation, but that comes with time, too (see http://drupal.org for example). With an open source competitor, many proprietary niche players can't cut it. Take, for example, many of the competitors to SugarCRM (http://sugarcrm.com) - Microsoft tried to buy them because they couldn't figure out how to compete - I believe they now own a substantial share.

OpenOffice is every bit as usable as MS Office, and it even uses a proper open standard file format (ODF) to boot. Entire school systems worldwide are switching to it, as are governments. It's changing the entire landscape for Microsoft.

At the moment, Microsoft are embarrassing themselves by trying to get their proprietary format, OOXML (which they made "open" by adding the word "Open" to the name. It's not really open in any other way) accepted by ISO so that they, too, can meet nation government requirements for an "open standard" file format. The NZ open source community, among others, is working rather tirelessly to see that it doesn't happen.

When you think that a) the internet wouldn't exist without open source, and b) more people develop with open source languages like Java, C, C++, Perl, Python, PHP, Ruby, Javascript, etc. than with VB, C#, and .Net, and c) Google, IBM, HP, Sun, Oracle, etc. are all getting behind Linux (Google uses it on its entire search infrastructure of several hundred thousand commodity machines - in fact, if I'm not mistaken, this Blogger site runs on GFE, aka Google Front End, which is a Linux-based webserver), and d) the fact that Vista, Microsoft's great hope, is substantially underwhelming, I think things are looking pretty rough for the "standard" as you described them...

But, then again, I guess you could say I'm biased: I only bet my business and entire career on open source.

Kind regards,

Dave

David said...

A quick digression: On the topic of MYOB, perhaps you can help me. I had an interesting experience just before Christmas. I received two emails in quick succession with this subject: "MYOB has the solution for your growing business".

They were sent by the marketing department at MYOB to two separate email addresses I use. I never requested to be on the MYOB marketing mailing list, and didn't know why I'd been sent these messages. I responded to Lara Fleming at MYOB asking that I be removed from the list.

MYOB email's subject line is patently incorrect. MYOB does not, in fact, have a solution for my business. My business has made the business decision to run Linux. MYOB's flagship software does not, to my knowledge, run on Linux. Therefore, my business cannot run MYOB, and it cannot, therefore, solve any problem my business might (or, in this case, might not) have. I respectfully recommend that, if you have any influence in the marketing department, these points are carefully considered before future mail-outs - particularly those which are unsolicited and false in their information - are sent.

Also, I'm quite interested in knowing where MYOB found/purchased my email addresses for the purpose of spamming me...

Jono said...

Hmmmm - Apache - anyone know of commerical websites hosted on Apache?

As for business continuity - MYOB dropped support for Win 9x in 2007 because Microsoft dropped support - despite the many customers productively working away on Win 9x. So who is MYOB being loyal to?

Matthew said...

I live and work in a small(ish) town of about 70,000 people. When I came here about four years ago, mine was the only free software/open source IT business. Today we're building up a small network of them.

I can tell you that the only way proprietary IT businesses in my town are viable, and their products and services affordable to the local businesses, is prolific copyright violation. Yes, uninformed people like the "Microsoft" brand, especially if they don't have to pay for it.

Had I tried to (legitimately) use/sell/support proprietary software, my business wouldn't have lasted a year. I lose a lot of work because when I tell people I don't don't use, sell or support non-free software they run for the hills, but the work I get pays okay, doesn't break any laws, and crucially doesn't compromise my clients' freedom.

Proprietary software still has the upper hand in certain niches (such as accounting), where there have been opportunities to invest a lot of money on development upfront to get a product out the door before anyone else and corner the market. The trade-off is that the initial investor demands a monopoly on distribution and control over the development process in perpetuity. This is not in the long-term interests of the software's users.

There may always be some niches where this business model is viable for at least a little while, but even where the software is initially technically very good, it tends to rot under monolithic control. I think in recent years the evidence has been overwhelming that proprietary software is going to become more marginal. Microsoft can't even sell Windows anymore, for heaven's sake.

I love Drupal, and I think in time it's going to become, as many are predicting, "the Linux of the web". My prejudices, notwithstanding, whether the MYOB-killer is built on Drupal or something else, I think the clock is ticking for MYOB. The only way to avoid this is to do a Sun-like turnaround and relicense the software in order to retain and expand the services part of their revenue.

Stuart Bale said...

To everyone - When Bevan told me he would let others know about my post, I never anticipated such a quick response to my little hidden away blog, although I guess I did ask for comment.
Please give me a little time to digest this and reply accordingly.
Please also be aware that, even though I am an employee of MYOB, my comments are my own opinion, and should not be construed as the opinion of MYOB.

Stuart Bale said...

David (re: a quick digression ...):
If you would like to email me directly, I am happy to investigate and provide an answer to your questions around how MYOB came to email you.
My email address is stuart.bale@myob.com.

Stuart Bale said...

Jim:
I agree with you ... perhaps I didn't set the context of my entry appropriately. By referring to Bevans comments about using an Open Source platform (Drupal) rather than Microsoft.Net, it might be seen that I support the opposite. In fact, David has posted that I have said this, when in fact I haven't.
I just want to clarify: My point is that I don't believe that Open Source solutions today are commercially viable.
I am not excluding that in niche situations, they may be relevant, such as in schools, where they often have no budget and so need free software (e.g. OpenOffice), and perhaps justify it under the guise of the need for 'Open Standards'. I would be intrigued to know how many users of OpenOffice chose to save the files in Word format - I guess that is something we'll just never know.
Nevertheless, my point is around commercial viability, which has somehow been lost in the depth of these comments (including my own). Jim, as you suggest, perhaps I should ask what people think Commercial Viability actually means?

Stuart Bale said...

Jono, I'm not sure if you have a sense of humour, so please don't be offended when I ask if there has ever been anyone who has worked 'productively' on Windows 9X ;-).

Jim said...

Actually I meant for you to define what you thought "Commercial Viability" meant :-)

I hope that the other commenters have been able to show that there are valid examples of businesses *using* open source being commercially viable (Google makes money) and of businesses *providing* open source development and support (David and Matthew show that). Perhaps MySQL fits both categories nicely?

There is no single "Open Source Community" to exist as an entity, that intends to demonstrate any "Commercial Viability". Equally, there is no "Closed Source Community" to show you what a sample EULA should look like, and to provide a cost vs complexity matrix. So perhaps you'll have to be a little unconvinced in that regard ...

"Viable" as a word in this context should mean "Capable of success or continuing effectiveness; practicable: a viable plan; a viable national economy.". Open Source is surely technically viable in all senses of that.

I think you're using "Commercial" in the sense of "Having profit as a chief aim: a commercial book, not a scholarly tome.". Open Source does is just software -- it doesn't have goals and aims of itself -- that remains an attribute of the author.

(Thanks to Answers.com / American Heritage dictionary for those definitions)

So, under that light, for an example of Commercially Viable Open Source you're looking for something produced primarily for profit, and that is successful and functional.

That sounds like Google, MySQL, Red Hat ...

Zedcar said...

Hi Stu. Interesting thread to start the new year... But sorry, I think you have asked the wrong question. (In so saying I'm reminded of one of my favourite chick flicks 'Just Like Heaven' where the dude keeps saying "righteous", but I digress...)

The point about open source is that it is not designed to have commercial viability in and of itself. It is not intended to be 'free' (as in beer) but free as in 'freedom'. The intent is to give developers the freedom to use the best and most suitable tool for the job, and freedom for users to avoid getting locked into proprietary file formats. Products & services developed under typical open source license restrictions can be sold just like anything else if they can find a market.

Anything has commercial viability if it solves a market need more cost effectively than the alternative. A segment of a market will determine whether the thing has value.

So the question is not whether Open Source has commercial viability. The question is whether proprietary source can compete in the long term. Of course it can't - as along as developers focus on meeting market needs.

huntdesign said...

Stuart

I think you ask a great question. I think you are demonstrably wrong in your conclusion.

My business is based on open-source software. My business is profitable. Thus open source is commercially viable. Is there anything else that needs to be said?

Are you suggesting that Google, Yahoo, MySQL, Ubuntu, RedHat and all the other open source vendors are not commercially viable?

You seem blinded by the fact that open source is free (as in beer). You seem to think that this makes it viable only when no commercial alternatives exist or for customers that have no money. I am quite happy for you to continue thinking this way, but for the benefit of others I'll point out that my business (Drupal-based web development) has an advantage over proprietary CMSs in several areas: no vendor lock-in, no data lock-in, no license fees and a worldwide developer community of thousands (that's more than most commercial CMSs will ever muster).

Interesting that you pick on 'assurance on continuity' as a positive aspect of commercial software. The facts are that numerous commercial enterprises have gone out of business and taken their software products with them. Just ask customers of some these companies: http://en.wikipedia.org/wiki/Category:Defunct_software_companies

Since you mentioned 'assurance on continuity', can you detail the source code escrow arrangements MYOB has made in the event it is no longer in business? The beauty of open source is that code escrow is built in.

BTW, where do you get off accusing Bevan of bias? Are you saying that your employment at MYOB hasn't coloured your thinking?

Tino said...

From a technical perspective the comparison between .Net and Drupal is an invalid one.

Drupal is a web application framework which sits on top of PHP while the .Net Framework is much larger than a programming language or application framework. See Wikipedia .Net Definition for more info.

Maybe a comparison could have been made between Drupal and the Ironspeed framework which is targeted at .net.

My point is the comparison that was made about .Net and Drupal was not an apples against apples comparison.

- With regards to Linux, I was pleasantly surprised by how nice it was to use but honestly speaking there didnt seem to be too many options when it comes to usable programs that run on it.

Also if you are worried about switching between versions of Windows wait till you find out how many flavours of Linux there are out there.

How many of you linux Developers have a product that you support on the different flavours of Linux. Multiply the different flavours by the different versions of each flavour and in my opinion you end up with a testing and support nightmare, no wonder the guy on Windows 9x doesnt isnt considering the move to Linux.

From a developers perspective (based on my own experiences):

-Not very backward compatible. I tried working with different versions of MySQL (mainly 4.x versions) and I had to implement a few work arounds to get things to work. It wasnt a very nice experience.

-Not up to date. I have always heard people talk about open source software leading the way in terms of innovation but then I was very surprised to see that MySQL only introduced stored procedures in version 5.0. For those who dont know the latest stable version fo the product is 5.0.51.

- On the other hand the free MSDE or SQLEXpress seems to be more stable and seems to have more features than MYSQL, although if you need a large DB you will have to pay for it.

-At about the same time that I tried using MySQL, I aslo tried using PHP. The thing is PHP is a great server side language and I know there are a lot of platforms that are designed for it. Also it produces nice functional websites like facebook...but...and yes there is a but, it is not easy to work with.

I tried using several editors and it looked liked most of them simply provided syntax high-lighting. I found it very diffuclt to debug PHP projects.

-PHP,much like MS Access is easy for anyone with little programming experience to get started but it can very easily end up in a spaghetti code project which is ultimately very difficult and expensive to maintain.

- In short, if anyone can show me a PHP editor that is comparable to the MS Visual Studio editor for .Net. I will gladly take a look at it but so fare I havent found any.

- The only thing I like about open source software is it keeps companies like MS on its toes otherwise to be honest I am not really too concerend about having the source code to an OS.

As for the viability of Open Source software...I am not so sure if it is viable especially considering the point about the linux maintenance nightmare.

Lastly, if you are after an MS response to the opensource community then try checking out the following web site, Codeplex.

goldboy said...

Stuart Bale said...
Of course, having said this, I think we are on the brink of a major revolution, with Internet connectivity and adoption rapidly increasing, that will see a significant change in the way people use computerised systems.

If this had of been said 5-10 years ago I'd have completely agreed with you. Hearing this today though...

Stuart Bale said...

huntdesign:
"...my business ... has an advantage over proprietary CMSs in several areas: no vendor lock-in, no data lock-in, no license fees and a worldwide developer community of thousands ..." in fact proves my point, in that none of your customers are locked in to YOU, so you could go out of business tomorrow. How is that commercially viable?
You say: "... My business is based on open-source software. My business is profitable. Thus open source is commercially viable ..." That is like saying, "A duck floats on water, wood floats on water, so a duck is made from wood."
Please note that the purpose of this blog is not for me to comment on MYOB - however if you seriously want to know this information because you use MYOB software, then I am happy for you to email me at stuart.bale@myob.com, and I will forward it on to the relevant people in order to get an answer for you.
Lastly, I don't deny bias on my part. I actually said of Bevan: "He believes (perhaps somewhat biasedly) that it would have been easier to build using the open source tool Drupal." I contacted Bevan directly to confirm that he was ok with my post - I meant no offense to him, and am willing to modify it if he is offended.
As i replied to Laura: "... my thoughts are indeed based around my experience, just as yours are..." That's why we are all different. <Rhetorical Question> Surely I could equally 'accuse' you of bias due to your background, couldn't I? </Rhetorical Question>

Matthew said...

Stuart Bale said...
huntdesign:
"...my business ... has an advantage over proprietary CMSs in several areas: no vendor lock-in, no data lock-in, no license fees and a worldwide developer community of thousands ..." in fact proves my point, in that none of your customers are locked in to YOU, so you could go out of business tomorrow. How is that commercially viable?

Uh, by doing good work so they keep coming back? Surely being able to guarantee that you won't screw your customers with vendor lock-in is going to attract customers?

Or to put it another way, as free software businesses become more commonplace, the prospect of vendor lock-in is going to become a very effective customer repellent.

goldboy said...

huntdesign:
"...my business ... has an advantage over proprietary CMSs in several areas: no vendor lock-in, no data lock-in, no license fees and a worldwide developer community of thousands ..."
Stuart Bale:
in fact proves my point, in that none of your customers are locked in to YOU, so you could go out of business tomorrow. How is that commercially viable?

Personally I would prefer to deal with this sort of business. You are absolutely correct in saying that their clients aren't locked into them which suggests that if they are still in business, and have been for quite a while, then they has at least 2 things in their advantage;
1) They must know their area
2) They must have good customer relations skills.

I'd much rather deal with someone that has my best interests in mind than someone that is wanting to hold my business ransom with vendor lockin.

After all, that is what you seem to be promoting with your comment above.

Bevan said...

Cross-posting related discussion: http://drupal.geek.nz/node/80#comment-98

> Where's Bill Gates or Steve Ballmer or Steve Jobs or Larry Ellison when you need them.

Bill Gates:
Resigning, because he doesn't want to take the blame when Microsoft implodes because it can't compete with the open source methodology
Steve Ballmer:
Still ignorant
Steve Jobs:
Busy adopting open source (like MS, MYOB and many others should have done too)
Larry Elison:
Who's that?

Seriously though, I had no idea that folk would respond like that. I haven't been following the discussion, but I hope that it is eye-opening for all involved. And I hope there are no hard feelings. :)

Stuart Bale said...

Well, I'm glad to see that there is still the same passion for Open Source today as there was fourteen years ago when I first started in this industry. And I'm glad to see that the same arguments are still being used too. Actually, Jim has convinced me the most to consider re-evaluating my opinion. Apart from his wild diversion on Word 2.0, after all: a) who would still be so stupid to have important information stored in that format; and b) what guarentee is there today that an Open Source/Open Standard equivalent will still be accessible in 18 years time?, I think Jim has managed to make some valid points.

I'm going to move on now. Please feel free to continue to comment, and I'll watch with interest too.

My next blog entry hopefully won't be on such a controversial topic, or at least not the same topic, however I am keen to come back to readdress the growing impact of the Internet on traditional software development, and hope that those that have contributed their comments on this topic might take interest in that too.

Jim said...

Thinking about real examples is much more fun than "I'm better!","No, *I'M* better" :-) I've been enjoying this discussion!

But, Word 2.0 ... it's an extreme example. MS have recently provided a much more timely example ...

http://www.theregister.co.uk/2008/01/02/xp_service_pack/
http://www.theregister.co.uk/2008/01/05/ms_office_sp3_woes/

From the first article ...
"The [Service Pack 3 for Office 2003] will block access to files including some Office Excel 2003, MS Office PowerPoint 2003 and MS Office Word 2003"

So, Office 2003 SP3 was unable to open files created by Office 2003 pre-SP3. The user didn't get an informed choice. The vendor made a change that denied the user access to their own (recent) data.

How far back would I have to go in producing original documents (invoices?) for compliance with IR? If there were no paper copies and I was audited? If I no longer had or could purchase a valid license for Office 2003?

Your last question is very simple to answer ...
"what guarentee is there today that an Open Source/Open Standard equivalent will still be accessible in 18 years time?"

None. But if your business were to rely on an 18-year-old standard, it has the legal right to preserve the source code in its own archives, and to hire a programmer to get it running again, if necessary.

(At the beginning of the year, Abdel Benamrouche announced that he had modified the original Linux release 0.01 to work with a modern compiler toolchain -
http://kerneltrap.org/Linux/Dusting_Off_the_0.01_Kernel
thus resurrecting an piece of code from September 1991 - >16 years old)

David said...

@Tino,

Sorry, Tino, the only thing in your post with which I can agree is that .Net and Drupal is not an apples-to-apples comparison. With regard to the rest, I'm still trying to figure you why you bothered to post the rest of your comment, given that you're disparaging something about which you clearly know so little about.

The only thing the rest of your response demonstrates is that you're not very good at researching stuff on the net. You'll find that your conclusions on Linux, its compatibility, up-to-date-ness, development environments are rather one-sided and most experienced Linux developers would probably chuckle and shake their heads in mild disbelief at your naivete.

Linux and open source easily surpass all of the shortcomings you list in your missive - but you'd have to be a bit more tenacious in your exploration to discover that. You'd probably also benefit from broadening your sample size a wee bit beyond MySQL and PHP to include some of the hundred thousand other open source projects out there before jumping to those conclusions.

On the point about "not being up-to-date", I'm amazed that you'd make an assertion like that. Linux and other open source projects surpass most commercial development efforts in terms of timeliness. Ubuntu Linux, for example, made 7 major releases in less than half the time it took Microsoft to go from XP to Vista - and it's fascinating to hear our hardware resellers telling us about all the returned Vista-preinstalled machines requesting an "upgrade" to Windows XP (for those cases where the customer didn't migrate to Linux, instead)...

If you want to learn some more about open source and why your perception could use a bit of updating, feel free to contact me through http://davelane.name.

Dave

Stuart Bale said...

Dave:
You've brought me back!
Thank you for the link to your own blog - it meant that I came across this absolute gem there. I must say that your banter with poor old Brian Jones of Microsoft made me laugh out loud! In particular, your suggestions as to how Microsoft can change its approach. I strongly recommend any reader of this blog to take a look solely at Dave's summary of the conversation ... don't bother with Brians replies ... as an M$ employee I guess he has to behave.

Stuart

David said...

Hi Stuart,

I'm blushing. And I'm honoured that you'd stop in for a read :) Yes, Brian's in a fairly thankless position acting as the lightening rod for a company that's trying to appear to be something it most certainly is not: worthy of respect and trust. Microsoft's OOXML effort is so cynical and deeply flawed, it's almost impossible to believe that they're giving it a go... Clearly, they're trusting that most of the people involved in the ISO and voting nations' standards bodies are as smart as a box-o-hammers. Sadly, I think they're sorely mistaken.

Cheers,

Dave

Bevan said...

This is probably just flamebait, but here's an interesting related analysis of acquia, a prime case example of open source commercial viability. Note also Jeff's work in Adobe where his job was essentially to embrace open source: http://www.redmonk.com/cote/2008/01/08/acquia-a-commercial-company-for-drupal/

Tino said...

Dave, I will give you top marks for being passionate about open source:-)

Here is a link that contains a list of the different flavours of Linux,Wikipedia Linux flavours , (note that this doesnt include the list of different editions or versions).

So my question to you is, if one was developing a desktop app targeting the Linux platform, would it simply work on all of them???

How does the compatibility between versions and flavours work?

Seeing as I am not a Linux guru like you, I dont understand how you can avoid the testing nightmare that I mentioned in my previous post, however I am willing to be educated.

I am after a factual response rather than a passionate dismissal of my question:-)

Jim said...

I hope Stuart doesn't mind us digressing a bit here :-)

Tino, you are quite correct that the Linux distribution landscape is very complex to an outsider. However, if *as a business*, you were wanting to target a desktop application to "Linux", you would need to do a bit of research.

People like Google and Opera have all released "commercial" desktop applications to the Linux market, so you could do worse than to look at what variations they are willing to support ...

http://www.google.com/linuxrepositories/
http://www.opera.com/download/index.dml?custom=yes

These guys basically agree that the current Ubuntu, Debian, Fedora, Mandriva and SuSE distributions are the 'accepted' common deployments.

In the corporate enterprise server market you'd be looking at something like Red Hat or SuSE -- but you would have to come to that conclusion by studying your target markerplace customers, rather than just general Internet usage of Linux.

Dave I'm sure will answer soon, as he is something of a specialist in cross-platform GUI development strategies -- where you can use a single common code-base to develop desktop applications that will run on Linux just as happily as Windows and other things.

David said...

Hi Tino,

I can see why - particularly if you've come from the MS world - the number of Linux distributions would leave you with the initial impression that it would be hard to target Linux as a software vendor. But that's not necessarily the case. There are a few reasons which I'll cover below:
1. Vendor-independent open standards
2. Incentives
3. Clean interfaces
4. Modular design (quality code)

I'm aware that there are hundreds of Linux distributions. They fall into roughly three camps:
1. from source code, (e.g. gentoo, slackware)
2. RPM-based, (Redhat, Fedora, Mandriva, PCLinuxOS, SuSE),
3. Deb-based, (Debian, Ubuntu, Mepis, Xandros)

There is, however, something that unites just about all of them: LSB (the Linux Standards Base) see http://www.linux-foundation.org/en/LSB

This is a distribution independent specification for what a linux is. As far as I know, all of the major Linux distros adhere to LSB. More importantly, Linux distributions allow introspection - there are known tools on any Linux system which allow software to configure themselves to match the requirements of the given distribution. These are available on every Linux system for any software distributor who wants to target multiple Linux distributions. As Jim rightly points out, clearly some software vendors - e.g. Google, Mozilla, and OpenOffice.org, etc. are already doing this. Ask them which operating system is easier to target - Linux or MS Windows...

Another thing: Linux distribution providers are keen to see good software made available for their distributions. Often, if a software vendor initially targets one of the major Linux distributions - particularly if they make their software available in source code form with a palatable license - other Linux distributors have a strong incentive to do the work to make that application can easily installable by users of their distribution... this removes the problem for vendors almost entirely. As a user of Ubuntu Linux, I see this process in action every day.

Final thing to consider: you call Linux "a testing and support nightmare". This, I presume, from the perspective of someone who has never tried to distribute software for Linux. How many variants of MS Windows do you think there are currently in the wild?
see http://en.wikipedia.org/wiki/Microsoft_Windows
There are quite a few - off the top of my head: Dos (yes, there are still commercial Dos implementations in daily use), Windows 95, Windows 98, Windows 98 SE, Windows CE, Windows ME, Windows NT (hey look! Windows CEMENT! ;) ) Windows CE Mobile, Windows 2000, Windows 2003 SBS, Windows Media Server, Windows XP Home, Windows XP Pro, Windows XP Home SP1, Windows XP Pro SP1, Windows XP Home SP2, Windows XP Pro SP2, Windows XP Home SP3, Windows XP Pro SP3, Windows Vista Home + 6 more variants, plus multiply that by SP1... you get the picture. The illusion that MS Windows is a single platform is very... illusionary.

As for modularity... if a Linux vendor can get their software working with a range of versions of underlying dependencies, then they only need to target RPM and Deb package management systems (with their well documented open formats and dependency listing mechanisms) to create software that can be distributed on nearly any Linux. When I lived in Seattle in the mid 90s, one of my best mates, a guy called Davi Kutz (for the record, he'd graduated from the University of Washington as the top engineering student of his year out of over a thousand) was working at Microsoft. His full time job was testing the MS Office 98 workgroup "ticker tape" feature... the one that allows a boss to float a ticker tape-like message across the top of the window of each employee's Word to alert them that they're fired, or have to work late, or whatever. Davi worked 80 hours a week going through every possible permutation of the ticker tape widget and checking for conflicts wiht every other possible MS application, option, feature, etc. and a bunch of third party apps (Adobe apps, Intuit apps, etc.). Davi, as I've already alluded, is a smart guy. He knew that he only had to test these things because - get this - Microsoft wasn't even adhering to its own internal interfaces. Its code was a MESS. No modularity, no consistence, poor internal documentation. Talk about spaghetti. Imagine writing code to target that platform! A nightmare. Davi quit shortly afterward, realising that all of his efforts were going into putting a fine polish on a piece of turd...

He know that Office 98 was going to get released despite the fact that he, himself, had submitted hundreds of bug reports to overworked devs who'd never sort them out in time for the "gold" release... it was soul destroying.

Linux makes no effort to hide the variability in various distributions - that's with standards, clean interfaces, sound code, and proper modularity... it doesn't matter. Microsoft promotes the impression that it delivers a consistent, seamless product line... Yep, it's integrated all right: it's well and truly bound together with thin strings of dough... Microsoft offers a monolithic face to the public, but under the bonnet, it's a different story.

That, by the way, is the main reason Microsoft would never open its Windows source: sheer embarrassment.

Cheers,

Dave

huntdesign said...

huntdesign:
"...my business ... has an advantage over proprietary CMSs in several areas: no vendor lock-in, no data lock-in, no license fees and a worldwide developer community of thousands ..."
in fact proves my point, in that none of your customers are locked in to YOU, so you could go out of business tomorrow. How is that commercially viable?
You need to understand that there are more ways to be commercially viable than by requiring coercive lock-in. My "lock-in" comes from doing good work at a good price. It's based on old-fashioned values like mutual respect, integrity, honesty, credibility and professionalism. I like the fact that customers come to me year after year because they want to rather than becaus they have no other choice. Open source facilitates this type of relationship between vendors and customers, and thats why I think it will ultimately win the day over most proprietary approaches.

Trevor said...

Hi everyone, I have been reading the comments with interest over the last few days. I know Stuart outside the virtual World so hopefully he will still shout me a coffee occasionally after my comments. I'm with the "Yay for Opensource camp" I think some of the difference of opinion comes from different world views that opensource / closed source advocates have. I don't think many people will become billionaires from opensource but large amount of hardworking programmers and small to moderate businesses can make a good living from it supplying good services.

From a consumers and users point of view I would be stuck without opensource products. I use a great IM called AdiumX because Microsoft supplies such a poor version of MSN for the Mac. Another great app I use and couldn't do without, Quicksilver, recently opened up it's code (I am not sure if it is officially opensource) to keep it alive. Interestingly although the program is a gem the code produced by this on man team is a bit of a mess and enthusiastic users are now getting involved to tidy it up and getting it to run much more efficiently.
I think detractors of opensource ask how they could make money from people like me through opensource projects. The answer is the don't. It appears to me that money is made from small to medium sized business who employ IT professionals to set up their business with opensource solutions for a variety of reasons or to adapt an opensource program for their specific needs that are too unique for a commercial vendor to have considered.

Opensource also has another objective. A social one. Corporate entities are about making money. I have no problem with that, I would like a bit of that myself, but their concern for you as a consumer only lasts as long as your wallet is open.

Oh sh*t, I think I have written a blog instead of a comment. woops, sorry

David said...

Hi Trevor,

Yes, luckily quite a few people in the world are not primarily motivated by money. Quite a few people (often open source software developers) are more motivated - if they have enough money - by producing great software and by living a balanced, healthy, ethical life. I'd count myself in that bunch. I don't know what I'd do with 10s of millions of dollars, or billions. Frankly, I think those sorts of disparities in wealth are bordering on criminal - those disparities point to major imbalances in the economic equation, which means someone is taking advantage of the system... that's what proprietary lock-in does. I suspect that someday it'll be illegal (although it might need to be legislated - perhaps market forces will just reject it).

So I think you're right - there will not be many more software moguls - but there will be very well paid, valuable, successful software developers building open source solutions.

I actually think that open source is the mechanism for the economic "adjustment" in the market to correct the inequities that now exist due to the peculiarities of proprietary software and the associated lock-in.

I don't see any inherent reason why a good IT person should be paid more than a good mechanic, craftsman, doctor, writer, etc. All are skilled trades requiring more than mere competence - they require understanding, ethics, community awareness - and, ultimately, all are service providers.

I, for one, am happy to earn a good living plying my craft - IT consulting, software support, system administration, and software development. If I can build a business which represents a real value to society, bully for me! I might even be relatively financially successful, but I would only want that success if I knew I wasn't doing it at the expense of goodwill from my customers, or by compromising my ethics.

Others in the software industry seem more than happy to do both.

The question most closed-source people will ask is "if everyone's doing service, where will the great software come from?" All I can say to that is... where does it come from now? Not Microsoft, that's for sure. It comes from small firms writing applications for customers. In my experience (as director of a company that develops open source software), it doesn't matter much to most customers whether the software is closed or open source. The continuity argument is one of the best reasons for a customer to insist that code written for them is open source - it takes a huge risk out of long term maintenance. It also allows much easier propagation of new software to market - if any end user can simply install a full working application where they need it (rather than having to go through a process like demo, trial period, budgeting clearance, sign off from accounting, arranging the purchase agreement with the proprietary vendor, determining number of seat licenses or some bulk purchasing agreement, upgrade contracts, long term support agreements, etc.) then I think the open source app will propagate orders of magnitude faster. Witness the internet - it's open source - Microsoft and other proprietary vendors are bit players in that market... and the current momentum is certainly away from the desktop towards the "cloud" so I know where I'd be looking for investment opportunities (not towards Redmond, that's for sure).

Cheers,

Dave

Trevor said...

Thanks Dave. Although my heart is definitely with your augments in general I have to respond to a couple of things. Firstly I didn't say that there would be fewer software moguls in the future. Infact I think there is still plenty of time left for some to become immorally rich. The number of Chinese and Indian billionaires is on the increase and many of these will be owners of software companies.

I think there will always be Apples, Microsofts, and MYOBs. If you are reasonably intelligent and competent you can normally achieve the goals you set for yourself, and as corporations have one over riding goal, to make money, that is what they will achieve. And they will do it better than community minded opensource companies who are distracted by other issues such as fairness and good will.

I don't think we should legislate to prevent this either. Opensourse is about free as in freedom after all. To truly stand for freedom you have to defend those who wish to take advantage of that freedom too. After all if we define what is allowed and what is not through legislation aren't we curtailing freedom? Obviously absolute freedom is not possible and must give way to order and restriction at some point for society to function but we must be careful where we draw that line.

I don't see this as a war that must have one party vanquishing the other. Both have a future. It would be nice if they could play better together. More mutual respect. As a Mac user I am aware that Apple has used opensource software for projects such as Safari and OS X. They have done well from opensouce and I hope they are generous with what they return to the community.

The other trouble for free thinking, slightly anarchistic opensource people is that they are a different breed to the majority. Society always has had and always will have a majority of people who succumb to the herd mentality. A few people will have the power/money, a bunch will think for themselves, often bemoaning the direction of society and majority will just follow the herd leaders not realising that what they do makes a difference while they are milked for everything that the rich and powerful can get out if them. And by and large most people are happy with this even if they are aware of what is going on.

Oh dear, I am letting my inner communist out now...I am going to say it anyway. I don't think the West can afford to let go of the model that allows MS and others to thrive. Software patents and copyrights are extremely important to Western economic dominance. Our wealth and power comes from the ownership of knowledge especially now that we have moved so much manufacturing power to developing countries. If knowledge was made free there would be bigger loosers than just MS.

David said...

Very interesting post, Trevor. Thanks for the response. With your last paragraph, you're definitely letting your "inner communist" show - although I'm not quite sure you meant it in the same way as I took it.

The West may not think it can afford to let go of the "artificial scarcity" model for intellectual monopoly (aka IP). That's because, as you say, it's been able to lock up knowledge.

Brief expository digression: When I say "artificial scarcity" I mean this (for the uninitiated): ideas (i.e. patents, copyright, etc.) are different from physical objects. If I give you my lawnmower or bbq, you have it, and I don't. Therefore, with property theft, I lose what you gain. With ideas on the other hand, if I publish my ideas or music (see http://davelane.name/music) or image (see http://davelane.name/gallery) or code, you can copy it, and you have it, but I still have it too... Unless, I can somehow impose an artificial limitation on your use of it... I say artificial because it has to be enforced by someone external, like the police/military/legal profession, etc. (at HUGE cost to society). End digression.

Today, even though the US (and I'm an American, by the way) owns heaps of patents, etc. you couldn't build a car or a bike or a toaster in the US anymore without depending on Chinese, Japanese, Indian, Malaysian, etc. components. America's self-proclaimed pre-eminence is at the whim of China, Japan, and Brasil (who will officially own the US sometime in the 2020s based on the rate at which they're underwriting US debt).

The idea of "intellectual property" is as communist a concept as you can get. It's a government imposed monopoly - enforced by rule of law limitation on the free market. Its our own slice of communism, where the government exerts absolute control over the market.

Make no mistake - Microsoft and most other major IP holders are fundamentally anti-free market (exceptions include crowds like this: http://www.patent-commons.org/) . Competing in a free market is their worst nightmare.

Open source, on the other hand, removes those artificial barriers - except for copyright - which depend on government. Open source believes in giving credit where it's due - in the physical representation of an idea (i.e. a computer file, written page, published artwork, etc.). Open source software is the closest thing to a free market that exists on earth - software succeeds based on its quality and fitness for purpose - NOT because it's the only software that's legally allowed to let people access data that they couldn't access in any other way.

So, for as long as the market tolerates it, proprietary software will exist (as funded by the "herd" you described, Trevor). That, like the lottery, could be thought of as a "special tax on people who are particularly bad at maths".

At some point, however, software will be like oil or coffee, and will be largely a commodity - each software package will work between various open standards. It's a natural transition - it's already happened in most other industries, e.g. machine screw thread sizes, automobile parts, arable crop seeds, fertiliser composition, etc. Would you buy an espresso maker that only allowed you to use PAMs High Octane Coffee Beans (TM) and burst into flames if you put C4 coffee beans into it instead?

The non-"West" (i.e. everywhere but the US) will continue to far exceed the "West" in its development, unhindered by an oppressive government system of artificial intellectual monopoly enforcement that the US and its lapdogs have imposed on their populations. Software - like industrial production - will thrive in those lesser controlled markets, and will soon surpass the head start gained by capital rich economies (the US) who were lucky to be physical resource rich at the start of the industrial revolution.

:)

Frankly, yes, you might call me anti-American - I'm a meritocrat. I believe that success and respect should be earned - tied to an exchange of something of value like physical things, skilled services, etc. - and not based on government assisted monopoly rents :)
Don't get me wrong - I'm not against residuals, particularly if they represent an exchange that provide continual value (e.g. a managed web hosting account) - that's good business. I am against residuals that are collected because a customer cannot find an alternative supplier due to lock-in (e.g. a "forced" upgrade to MS Office 2007 because 20% of your customers are using it (i.e. MS Office 2007 was the only option when they bought their new Dell computer), and its file format can't be read by MS Office 2003).

Right, back to work.

Dave

Zedcar said...

Wow. I've been away for a week and this debate continues. It's a fascinating thread, so thanks Stuart for kicking it off.

I guess I've always considered myself a 'free market capitalist' but from reading this thread I can see now that the phrase is an oxymoron. Like all things in society, we need to find a balance between freedom and control because the extremes of anarchy and communism are not acceptable to most of us.

Commercial viability comes ultimately from perceptions of value that in turn arise from the cost of substitutes. There will always be value in software of all types, but as Adam Smith taught us, all markets tend to equilibrium and that's why maybe the excessive returns of the past may no longer be seen in (controlled, closed) software.

It's interesting that something as mundane as software can become an exposition of such deep philosophical issues that have been with us since Adam & Eve.

But my question is this - how come no one agrees with me that 'Just Like Heaven' is one of the best chick flicks ever?

Zedcar said...

Here's maybe the best example so far of how 'open source' can be commercially viable in the mass market... the Asus Eee PC. See my blog for my first impressions.

I guess what's happening here is that the concept of embedded software is finally reaching the masses. The Asus Eee PC is not a PC - it's a fully functional, consumer ready internet appliance. The consumer doesn't need to know or care about the difference between software & hardware - you just switch it and start working. And it's selling like hot cakes.