Skip to content

Change is expensive

Change is expensive. Standards are hard to change and anything that is not completely open is brittle. Or, why I am hesitant to embrace anything beyond basic web or documents until we develop a proven track record.

What I am trying to do is to evangelise the importance of richly interactive text by talking and writing about it, hosting dialog and by ‘putting my money where my mouth is’ by developing software to demonstrate what I am talking about which is how I make it real.

What I am not trying to do is to remake the whole world which is what some of the great minds have done. We live in a world of electronic infrastructures which are real and exist and are established, much like there is always ground to build a house on–you never build a house in thin air, and changing users habits is very, very hard. This is why I am happy to piggy-back on PDF and WordPress/HTML and web standards such as Linked Data Notifications. We can implement technologies behind the scenes but I am very wary of anything that requires user setup. I am not here to serve the tech geeks, they can take care of themselves. I am here to work on systems for the general knowledge worker who wants to develop a deeper literacy with the systems they can use for their work, but who are more focused on their work than what’s under the hood to make their systems work.

I am also not trying to tie together commercial or emerging standards such as Hypothesis or dokieli (I have tried to find the history and what’s behind the name but I have not found much. This one annoyed me, including the name! But I have a sneaking suspicion part of our future will lie here…) since the work needs to fit into actual workflows–workflow chapter markers or end points or results or whatever we want to call the parts of workflows where there is a standard are very, very hard to change–so that is again why HTML and PDF are important though of course, they need to be improved upon radically.

My main project is Liquid | Author and it’s taken a lifetime of false starts to get here. It is my personal project and I will be the designer with the last word though I of course value and elicit feedback. I saw the General Magic documentary this week and it was both intensely moving and sad but what I remember from the time was that their work environment was too cutesy. I never bought nor wanted any of their products though I was amazed by their vision and the people working there. It may very well be that my own software is out of touch, but reviews by real users convinces me that it’s on the right track.

What I feel I have to focus on is polishing and polishing the actual end-user process of authoring in Author and reading in Reader, while supporting open standards for richer interactions and richer interoperability, hence Visual-Meta and Citation Meta (not sure why only one has a hyphen).

I am hugely excited to be working with the young and brilliant Raine and I’m sure she’ll push me into places I never thought about really, such as IPFS and other enabling infrastructures but she asked me–half jokingly I think–what I have learned since I was her age and my answer is this post:

1. Change is expensive so only change what really needs to be changed
2. Commercial services are not likely to last a thousand years, all companies go out of business
3. Respect the actual user workflow, I really have problems with PDF (it’s frozen!) and WordPress (over-reliance on professionally/paid themes) but you go to the market to sell, you don’t shout outside the castle walls.

And that’s it for this ramble. I’ve had my coffee and the light is coming up. Soon it’s gym and then this afternoon I’ll drive down to Southampton to see my advisor Les for the my PhD. Hopefully not much time will be spent on The Future of Text. And yesterday my beautiful baby boy Edgar turned 2.5. We went to the Extinction Rebellion, spent time with Chris…

Hegland, 2019.
…A McLuhan Quote and a call for Frugality (it was all quite moving)…

‘frugality’. Hegland, 2019.
…hung out at the RSA…

RSA. Hegland, 2019.
…and we buggy-ran through the rain. Perfect day.

Big Rain. Hegland, 2019.
We even saw a guy using a phone box (permission to take the pic was indeed received)!

phone. box. Hegland, 2019.
Best of all perhaps, was simply hanging out with a good book:

Reading. Hegland, 2019.
Happy 2.5 Birthday Edgar!

Published inCorrespondanceNotes On...Photography


  1. […] reply to “Change is expensive” by Frode Hegland. This is mostly sequential commentary in the order of paragraphs in the source. […]

  2. Where change is expensive, wouldn’t make it sense to work on making it cheaper? The example of Engelbart shows that change can be made sufficiently cheap so it becomes affordable to have lots of it. Who cares about actually changing standards (?) if you can have and make lots of them? Better to have standards than everybody doing their own thing in an incompatible, conflicting way! Isn’t the knowledge worker somebody with highly specialized needs, tools, methods, habits, training?

    PDF is fine for print and frozen rendering snapshots/images. WordPress and HTML are getting worse in terms of support for text, but with some effort can be done well enough. Wonder how a PDF reader or generator or a web browser comes onto the user’s machine if not via setup, except we’re in the days of passive, exploited users already who buy their computing devices as dedicated appliances, throwing it away and buying a new one if different software functionality is needed. That’s not technology by the way, it’s consumerism, and can’t do a thing.

    If there’s tech that’s supposed to help the knowledge worker (beyond his/her brain and sensory), I guess there’s no way around having geeks making it work, otherwise it simply won’t work and the knowledge worker is back at doing stuff as it was done since the beginning of time. Systems available for knowledge workers they can use to develop deeper literacy, a precondition is that these systems exist and were built by someone, and it depends on the how if they’re a good basis for enabling/powering the development of deeper literacy, in contrast to artificially or unnecessarily restricting it. But yeah, the knowledge worker doesn’t need to care much about what’s under the hood, if the tech that’s in there actually works well enough.

    Why care much about Hypothesis in particular, the open web standard is Web Annotation of course, why wouldn’t that be perfectly fine already to solve a whole range of problems? It’s far from perfect, I don’t like it too much, but at least one can work with/from it in comparison to have no standard at all. Just for completeness, there are other viable standards for the same thing as well (probably even interchangable/convertable), for example Ted Nelson’s EDL.

    Workflows are comprised of many individual steps, at each of which you can divert from standards or use different sets of standards. I don’t see any limitation with such an approach, to the contrary: workflows are a strategy for introducing more flexibility. Also, “workflows”, that’s plural. Note that there are some more other ways of how to talk and think about this, but they’re less prominent in some circles up to this point.

    As Liquid | Author is on the right track, if this tool does its job, great! There are just loads and loads of different jobs out there, so not everybody necessarily, exclusively cares about a single one in particular. I’m just personally wondering about the interoperability part, how could my stuff possibly interoperate with Liquid | Author, how could I possibly make it to? I guess that’s where standards come in, right?

    • Thanks. Change is expensive no matter what we do but absolutely, we should make it less expensive.

      I’ll be posting an update in a few mins. Please have a look and if you can comment as a blog post and not in the comments I think that would be really cool. :-)

      • Don’t get me wrong, I totally agree that it is currently way too expensive, be it the common lack of standards to bootstrap their semantics (not that it couldn’t be done today already, it’s just that nobody is really doing/trying) or the dominant model of building big, monolithic apps with loads of dependencies, which then only have internal interdependencies and subsequently can’t interoperate with other external capabilities.

        Posting as a blog post, you mean on

        • Agree with trying and that we should build smaller, interoperable pieces. Yes, jrnl or any blog :-)

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.