« Bono, from Davos | Main | In Davos, Cutting the Stench of Fear ... »



Let's just get a single well-defined document format for office applications without web support on the rails, shall we?

Trendy web-based editing hasn't arrived yet, but office formats are old and established and urgently needed ton ensure future access to a mountain of public and private documents.

It is important to ensure that proprietary formats (such as inherent in OOXML) are not allowed to find legitimacy in "standard" formats.

By the time that we have taken care of that, extension to CDF neend't be a difficult problem.

sp3cial 3d

"sysiphean ordeal for the blindered"

Dude, put the thesaurus down and relax...

Mohammad Khan

Interesting -
...and IBM got poor W3C CDF working group members to deny CDF's applicability to the office suite tools ...

What are you smoking Sam ? Did IBM had a gun to this guys head or had his wife kidnapped ? Someone doesn't support your fantasy so it must be IBM's fault. You sound so pathetic. Get a life.


Why don't you give is a rest.


Talk without implementation equals vaporware. Wake us when you have running code.

Sam Hiser


I'm about a few clicks less pathetic than IBM's 1997 fantasy called Lotus Symphony.

Sam Hiser


Good point. Will do.


I have a true story to tell Golodh.

Many years back, I use to use WordStar. After that, I switched to Word for DOS. Around 1996, I realized that even though I had the WordStar files on readable floppy disks, I didn't have WordStar anymore and my new version of Word didn't have a WordStar converter. As a result, I could not read those files, much of which contained valuable information (e.g. Documentation on the operation and maintenance of large electro-mechanical systems many of which are still in use today and will be around for another 50 years). This is the same issue that large corporations, government archives, and so forth are running into.

Printing these documents to paper, what Office formats are ultimately designed to do [why else do office suites present you with a virtual paper page when open a new document], was not very important. If I had wanted to clutter my office with paper, I would have printed the documents long ago. They were stored on floppies and tapes to save space, but would still be retrievable when required.

As a result of this issue, I began looking for an electronic document storage format that I could rely on which would still be readable in another decade or 2 or 3 or 5.
There was only one standard which I predicted would still be around and in use 50 years later. It was a relatively new standard in 1996 but it was clear it would live on long past not only my lifetime, but those of my grand-children. What made this standard so wonderful was its simplicity, its easy of use, and its versatility. That standard was HTML.

So from that point on, every important document I needed to store for long-term would either be in straight ASCII if it didn't need formatting, or would be in HTML if it did. Since an HTML2 plug-in for Word existed, I could use Word, receive word documents from my co-workers, but save everything as HTML.

Around 2000-2002, at work, an update from MS made our versions of Word unable to open any documents saved with previous versions of Word. MS has done that again with Office 2007. However, because all of my documents were stored in HTML, while everyone else was panicking, I was able to continue working. This simply re-affirmed my choice in using HTML (and being called stubborn about it).

Years ago, I read an article in a VMEbus trade magazine explaining why standards succeed and fail. The main thrust of the article was that new standards supplant old standards when the new ones add significantly greater functionality over the old ones (e.g. PCI's Plugn'Play over the VMEbus). OOXML adds no significantly greater functionality over earlier defacto office standards and so I think it'll be dead within a decade, likely with a lot of help from MS. ODF is a better standard since it's an amalgam of various other standards. But again, for straight documents, what does it really offer over HTML 5 with CSS3? Nothing significant. So it may live for 1.5 to 2 decades. 20 years is not "good enough". What about CDF (XHTML, SVG, SMIL and XForms)? We'll see. [If we called it HTML 6, that would be a different matter.] But HTML is here to stay and the sooner we make office suites that produce better HTML documents as easily as they produce ODF or DOC, the better.

[As an aside, when I began looking at this issue 2 decades ago, since I didn't want the contents of my document spread over different files (specifically those pictures we needed for documentation) where one might get lost when trying to copy them, I came up with a variant of HTML which would handle the task. I proposed altering area which could contain the picture data to allow faster loading (the HTML loader would work until it hit , display the document so far, and then start loading the pictures it found in . Either way, I think there are ways to improve HTML as a document format, and I think CDF is a good starting point.]

Sam Hiser


Thank you for your thoughtful [and time-consuming AND accurate] post.

J David Eisenberg

Without CSS 3, I don't see how you do headers/footers, complex page layout, etc. Is there a working implementation of CSS 3 that I've overlooked? Also, I'm not clear how you'd use HTML to mark up such things as indexable words. Is there some equivalent of ODF's text:change-start and text:change-end elements to track changes across a segment of HTML that by itself is not well-formed? These are important issues when representing a word-processing document.

The comments to this entry are closed.

Sam Hiser




This is a Flickr badge showing public photos from swhiser. Make your own badge here.

Locations of visitors to this page

Search PlexNex


View Sam Hiser's profile on LinkedIn

Powered by TypePad