Microsoft isn’t exactly known for it’s love affair with cross-platform technologies. If anything, they are the most anti-cross-platform technology provider there is. Windows and its eco-system is all that Microsoft is interested in; it’s where the bulk (nay all?) of its revenue is derived from.

Buying Skype for $8.5B is an interesting move. It’s roughly double what Google or Facebook were going to pay, if the rumours were true. Thus, Microsoft must believe that it will gain synergies (that horrible management/marketing word …) from integrating Skype technology (and brand?) into Windows in the Desktop, Server, Xbox and Windows Mobile world. It’s got a large fight on its hands with Cisco in the conferencing space and has been looking to make its Unified Communications solution really, really compelling. And maybe by adding Skype to the mix it will be compelling? But $8.5B compelling?

It’s hard to see how Microsoft is going to get its investment back. There could be some useful patents in the mix, and Microsoft is showing that it likes to throw patents around to interfere with markets and use legal tactics as competition rather that, say, just innovating and competing. However, keeping the technology away from Google et al. may also be part of there competitive strategy.

So there’s lots of interesting stuff in there. But for those of us who use Skype on Macs, Linux, Android, or anything that isn’t Windows, yesterday marks a turning point. Previous acquisitions by Microsoft that were cross-platform eventually became Windows only, with the other platforms left swinging in the breeze. What are we going to use instead? SIP, Google Voice, something else? Come on Google, sort out your Google Voice offering for the rest of the world …

He extols the virtues of using C# and mono (an implementation of .net) whilst disregarding any dissenting voices in the community.  Still, when he comes out with stuff like this:

I find Silverlight incredibly appealing – you get C#, you get a DLR (Dynamic Language Runtime), you get a fantastic graphics engine with a fantastic animation framework, you get video, you get audio, multi-language compatibility and so on and so forth.

Well, that’s great.  It’s lovely technology.  Shame it comes from a convicted monopolist that has repeatedly called Free and Open Software (FOSS) ‘cancer’ and keeps trotting out patent threats against the free software community.  So what’s the sensible thing to do?  That’s right build technology based on Microsoft ideas and try to get it into as much FOSS as possible.  Perhaps not.  But then he goes on:

The business side explanation is that we want to make sure that Linux remains a first class citizen on the web. As websites start using Silverlight we don’t want Linux to be in a position where you can’t access those websites.

Perhaps it would be better to not to use Silverlight in the first place.  This is like the whole ActiveX thing all over again.  This is why we want OPEN web standards not driven by ONE company.

Please read the rest of the article. One thing is very, very clear to me.  Miguel de Icaza seems to be several things:

• In love with all things Microsoft.  He even went for a job there; perhaps he did get it.
• Care’s not one jot about the damage that he might do to the community by validating and pushing Microsoft technologies into the FOSS community.

Still, the above is an ad hominem attack.  But, I have no respect for the man; I personally think that he is doing a very bad thing by developing Mono, by his blind desire to replicate .net, and with his completely naive attitude to Microsoft.  He wants to push mono in GNOME.  Which is a shame, because I like GNOME.  Still there is always KDE. Now, what can I use to replace TomBoy and F-Spot.

How many spreadsheet formats does Microsoft use internally for running their business on? Why should governments be denied choice in the same field that Microsoft itself exerts its right to chose?

Then there’s the specification itself. 6,000 pages at the first round, growing to something like 8,000 after the BRM. The sheer number of technical errors, failure to incorporate other ISO standards in the appropriate places, and downright contradictions should have ruled out the specification from becoming a standard. That Microsoft shoved this standard into the ISO Fast Track by getting ECMA to rubber stamp it shows a complete disregard for the standards process. Technical committees the world over agreed that the standard shouldn’t have been fast tracked.

Then there is the fact that it probably isn’t possible to implement OOXML. Microsoft Office 2007 doesn’t implement it contrary to what the popular IT press reports, and it’s unlikely to be implemented in Office 2008 (at least on Macs). So even Microsoft doesn’t think so.

So what it the point of this “standard”? Microsoft has expended vast amounts of time, and money, on countless countries’ technical review committees on a standard that it won’t implement. But perhaps it reveals the true purpose of why it did it. Microsoft certainly doesn’t want standards that other organisations can implement as it damages Microsoft’s potential for lock-in. By changing document formats with every version of Office, it can essentially dictate the upgrade cycle to the market and guarantee it’s office monopoly and revenue stream. By producing and getting standardised a fundamentally broken format it again gets to set the agenda for the document format. If we let them.

One oddity is that Microsoft repeats this mantra of choice. They say that customers should have a choice of document formats, and by standardising DIS 29500 (OOXML) they are simply providing customers with a choice. It seems reasonable on the face of it; shouldn’t customers have a choice of document standards?

One way to look at that question is to ask; shouldn’t there be a choice in electrical plug sockets? Shouldn’t there be a choice in the way the pedals are laid out in a car?

Each of these examples provides a clue to whether document formats should come in may different formats for the same type of document. There are many different appliances that plug into electrical sockets and there are a great variety of different cars.  In each case the customer benefits from having a single standard, but has the choice from many implementations of the standard. This is a good thing; multiple pedal layouts on cars would be bad from a safety and cost perspective. Manufacturers would have to support multiple standards so that customers would get the style they need; and certain cars would only come in one standard which would prevent some customers from being able to drive those models.

Multiple implementations of a single document standard are good for the customer. Plenty of choice, the ability to match features to price and the sure knowledge that the document will always be available. But it’s bad for Microsoft as it bypasses their lock-in.

Microsoft repeats that customers should have a choice and equivocates choice in standards with choice in products.  I want a single standard and multiple implementations to choose from. Then our documents can be used by everyone regardless of their product.

This has driven me to write about it. I think it is important for interested parties to take a stand, make a noise, and generally create a fuss about what should be open standards. This is the first of a series of posts where I state my case based on my own reading an understanding of the situation. But I will be boycotting all of the OOXML-like file formats, the .docx format and it’s stable-mates.

Why do I want an open standard?

Open, non discriminatory, standards are essential in ensuring that the documents we, and our governments, produce are actually readable and usable beyond the application that was used to create it. There’s no point in having “supa-dupa” document format “X” when you can’t even open it in 20 years time. By having a truly open standard with multiple applications being able to read and write the format, particularly if one of them is free software or open source, we can guarantee that we will always be able to read that standard.

However, if we continue to use proprietary, monopoly supplied document formats then we run the risk of not being able to access those documents in years to come. We run the risk of permanently losing these documents; this might not be too bad if it is just a letter to your Grandmother, but it is a disaster if it is a contract, historical paper, scientific research or other important documents.

My next post will deal with what’s actually wrong with the MS OOXML ‘standard’.

I naively thought that I would be able to use the web interface of a router made in 2001 on Firefox/Linux. Nope, just locked it up or crashed it completely². So, I then tried ies4linux. This uses wine to run Internet Explorer on Linux. That got me into the router but then the page showed all of these little empty boxes; a curse of Java applets. I installed the Windows version of Java (in Wine) but that just gave more errors. In desperation I loaded up a Windows 2000 image in QEMU I keep on the laptop (for just these occasions). But that failed too.

Why? Because the Sun JRE/JDK doesn’t have a class called DynamicTable. After googling for the error I found that two other people had run into the same issue; neither had solved it. And then it hit me – DynamicTable had to be in the Microsoft Java VM not the Sun one.

Why on earth did SMC use a browser interface that uses Java applets that only work with an obscure, out of date, Microsoft VM. Why on earth did they use Java applets? Does anyone else’s? I found an old copy of the Microsoft VM and installed it, but I already had the Sun one installed and they didn’t play fair. I’ve probably screwed up the Win2K image now. I gave up. At least the router has got a telnet/terminal interface. Naturally it’s not documented anywhere.

The Microsoft VM has all but disappeared, but those ISDN routers are still out there. This is what you get when you base your products around a proprietary technology that can just be EOLed on the whim of the company.

Telnet anyone?

¹Dial Up Networking; a faintly anachronistic term now that we tend to connect via WiFi or broadband.
²It’s a bit worrying that Firefox can still dump you out completely (thank god for auto-saved sessions), but then these pages were laden with Java applets.
³The router at least supports a Cisco-lite like terminal interface. Yes, it’s just as awful as the Cisco version.