No they're not because of multitude of reasons but they now use a open XML format that does way better interchange with openOffice etc. and don't actively try to shut down those products.
The keyword being now. It happened when they realised that they were going to be knocked off their pedestal if they didn’t open up - not when they were at the height of their power.
By the way, is the XML Office file format really open — to the extant where one can create a third party client which works as well as the official one without reverse engineering?
The demise had already started back then with the advent of Web apps and crappiness and delay of Vista. In fact, MS started losing its power and clout way back in early 2000s, even if the effects were not very evident until much later.
The year of the Linux on desktop/laptop is a fabulous myth - nothing more. Read the attached blog before posting sarcastic comments like this one.
On the other hand, Linux is pretty much everywhere other than desktop/laptop, which are becoming increasingly less relevant.
Perhaps demise is too strong a word, but decline from its status as the apex predator of the industry is more appropriate. MS is still powerful, but it is no longer the King.
Apple devices obviously don't run it, Android and ChromeOS might have Linux underneath but only expose Java, JavaScript, ANSI C/C++ lib and Android specific native APIs.
Android P could be released with a complete different kernel and only OEMs would notice, all user space apps would be kept running.
And that is where you are horribly mistaken. Typical Java/.Net coders may not care about the underlying platform while coding, but people working with large scale real world applications do. Java/.Net may abstract away the APIs, they still have to use the underlying stack. Their performance is still dependent on the performance of the underlying stack.
I have seen the same application with drastically different performance characteristics on different platforms.
Linux is the king on the server because it has been heavily fine tuned for the server over the ages. The cost, while a factor, is not as relevant. Any company deploying large scale applications on Linux has to hire enough engineers to keep it running that the cost of Windows is largely offset. If it were just a matter of cost, we’d see more deployments of FreeBSD and the likes.
It is the same reason people stick to Java despite all its shortcomings.
We don't see BSD deployments thanks MIT licenses, which allow for companies to keep their forks and the old AT&T suit.
The only reason UNIX shops migrated to Linux was to avoid paying for workstations and migrate to Linux.
Companies like SGI dropped their UNIX and are having Linux forks, as a means to outsource development costs.
Likewise with Oracle dumping Solaris and migrating whatever is relevant to Oracle Linux.
In 100% of projects I have been involved, Linux was always chosen as means to reduce license costs, as developers do 100% of their work on Windows workstations.
Well, I know Microsoft XML format has the word "Open" in its name. But is it open? Can other vendors achieve 100% compatibility with the official Office products' formats? I don't think so.
This comes up every once in a while. For a couple of years I worked on conversion filters for Word Perfect. The problem getting 100% compatibility has nothing at all to do with the "openness" of the file format. It has to do with the fact that file file format doesn't tell you how to render something -- the renderer does.
Just to give you an example, the file format tells you that there is a foot note. If the footnote doesn't fit on the page, then you have to paginate it (continue it on the next page). The file format doesn't tell you how or when to do that -- it just tells you that there is a footnote. Word happens to paginate footnotes incorrectly. Word Perfect paginates footnotes correctly. To import a Word document perfectly, somehow you have to break the pagination of the footnotes -- something that is practically impossible to do with a file importer.
Microsoft can't fix the pagination of footnotes because that's the way they've always done it. If they fixed the pagination, they would break decades of existing legal documents (which all have really long footnotes). They could add a new feature to paginate them correctly and tag them in the file format: For example "Here is a footnote. Paginate it correctly/incorrectly". I don't think they ever did that with footnotes, but they definitely did for a lot of other features (It's been over 10 years since I looked at the spec, so I can't give you a good example, unfortunately). Basically, you will see in the spec an option that says, "Render like Word 95", or something equivalent.
If you want to render Word documents perfectly (as well as your own), for each feature you need to be able to render it in your native style, and in style that Word has ever used -- even if these are fundamentally broken from a layout perspective. And then you have to add all of those Word styles to your file format, or else if you import it into your file format, you won't be able to save it.
Nobody does that, for obvious reasons. It's not Microsoft's fault. The opposite is exactly the same. That's why MS dug in their heels about not using Open Office file format -- it's an insane idea. I'm not a big fan of MS as a whole, but I interacted with the Word team a lot when I worked on conversions for Word Perfect. They were always extremely helpful -- to the point of sending me bug reports when I made a mistake on the conversions. They always knew that interoperability of their file format was only in their best interest and they did their best. Word is just a massive legacy system and they have their hands tied -- mainly because they always bend over backwards to maintain backwards compatibility for their customers.
I understand the feeling you are trying to get at, but in this case it is entirely misplaced.
Have you looked at the actual XML? Parts of it are just containers for binary data. How do you parse a tag that basically means base64-encoded Word6-style Compound Data? Good luck with that.
