A few semi-random points around my blogging and writing

I have a few points relating to my blogging and writing, sometimes more generally applicable. Since not all have sufficient mass individually, I publish them as a group:

1. When I first started my website, I had highly ambitious goals in terms of both quality of “markup” and continual improvement* of the texts themselves. Over time, the sheer amount of text has grown so large that I must consider these goals entirely unrealistic—attempting to keep them would cost far too much time, and would make writing a too boring leg-work task.

   *Largely based on my experiences as a software developer, especially with refactoring.

   Indeed, even the wish to make certain texts sufficiently high-quality for a first publication can cause severe delays—and I have often seen my self forced to draw a line and publish something three-quarters done just to avoid an eternal state of non-publication. (In cases where I felt that the importance of the topic to me, the amount of time already invested, or some other factor, made non-publication worse than sub-optimal publication.)

   Over-time, such complications have also changed my attitude towards blogging vs. running a “proper” website. While blogging is not optimal in terms of the resulting product, including the reduced ability to improve* texts, to link from an older work to a newer, and similar; it does have the advantage that it is easier to keep productivity up. A blog-like format is hard to avoid when the quantities of text grow beyond a certain point.

   *In the case of larger changes (arising e.g. through a better understanding of an issue, with a wish to expand, alter, retract, whatnot) a blogger will usually even have to resort to an entirely new post. While this too has advantages, it is more of a “necessary evil” than something positive, forcing me away from a more “book-like” result to a more “newspaper-like” one. However, I have noted that such larger changes on my website often led to problems with e.g. structuring or focus, and writing something new might sometimes have been the better idea.

2. Partially overlapping: What I put off for the future can be delayed by months, even years, or not be done at all. It is, for instance, quite common for me to mention an “upcoming” text and not actually write it until months later—or to write a text half-way and then to leave it be for a few months.

   Indeed, despite the aforementioned goals, my website still has many articles with TODOs or obvious defects in them, because I published knowing that I could (and assuming that I would) relatively soon make corresponding updates. Some of these have been around since the first year of my website. (2009! The time of writing is 2018…) Worse: My first major attempts at writing consisted of a number of (paper) notebooks, especially based on my experiences at the now defunct company Firstgate/ClickandBuy*—the hands down worst employer I have ever had. Most of the contents of these notebooks are still only present in the very same notebooks…**

   *Due to the “defunct” part and the long time gone past, I no longer have any hesitation in mentioning its name: Those reading my website might find references to “E4” (=> my 4th employer), which is an anonymized version of the same company. (The lack of a key to understand some such names is a good example of an “obvious defect”.)

   **It is uncertain whether they will ever be published: In addition to the problem I discuss here, it is quite possible that my opinions, priorities, whatnot, have changed too much in the intervening years. This especially since parts of these writings had a cathartic character. Despite my considering these notebooks the core of my writing for a number of years, it is conceivable that I will at some point simply put them in my shredder…

3. A special case of the first item is tagging and categorizations: As I have found over time, it is more-or-less futile to do such things manually, except on a very, very broad scale. This not just restricted to writing, but often in other areas too.

   Consider e.g. categories: If there are more than several categories, it is quite common that there is no single obvious match—implying that more than one category should be awarded, lest the readers look for something in one plausible category and miss it, because it is in another category. On the other hand, if this is done, we have the confusion that the same text (generally, “entity”) can appear in several categories. (This, in turn, might seem like a job for tags, but tags have their own problems, cf. below.) If the categories are fixed in number, there is often no really good match (implying “no category”); however, if new categories can be added to resolve this situation, then the number will tend to increase unduly, the risk of overlap is rises (because the new categories tend to have a less thought-through and ad-hoc character), and we also risk ending up with almost empty categories.

   Tags are usually* very similar to entirely ad-hoc categories, which are just thrown on various entities as seems fit, leading to complete chaos. To boot, we have questions like what degree of detail should be used, what number of tags applied, etc. Should e.g. an article on association football be tagged “association football”, “football”, and/or “soccer”? In most cases, only automatic tagging (and mechanisms with a similar purpose) make sense—to the point that I might even recommend not tagging most texts on the Internet at all, instead letting search engines and similar tools find relevant texts. I have even seen the recommendation to only use tags when the relevance of the tag is not clear to an automated tool from the text it self.**

   *Exceptions occur e.g. when the number of tags is small and/or their values are predictable. For instance, an email reader could use a few fix tags like “read”/“unread”, “urgent”, …; a version-control system could use tags indicating certain releases and other important events, and do so in unlimited numbers, as long as a consistent naming scheme is used; the window manager WMII, to which I have recently switched, uses a tagging system in lieu of “virtual desktops”, which works very well as long as the user does not do anything stupid.

   **Which obviously makes a mockery of tagging, because the most expected tags are then not set, and anyone who tries to use tags to e.g. browse contents will be lost.

   (Also see an excursion at the end.)

4. I have grown uncertain what to call my works: When I wrote mostly for my website, I usually used “article”; and I continued that use on WordPress too for a long time. Over time, I switched to using the word “post” on WordPress, seeing that this is the standard on blogs. For a few months, I have been torn between “article” and “post”, because I intend to return to my website in the long term, likely including some type of import of my WordPress blogs, which might make “post” misleading. Recently, I have resolved this by mostly speaking of “text”, which is more neutral, avoids the risk of being misleading, and also distances me from journalists*.

   *Recurrent readers will likely have noticed that I have a very low opinion of journalists—and I do not wish to be associated with them.

5. The “re-boot” of my website, which is one of the main reasons why I have taken a sabbatical, is likely to be one of the many things delayed, for the simple reason that there is much, including the above, that I want to think through before I start. I suspect, however, that the result will be something more like a blog* than the old website (cf. above); albeit, with better support for later changes, notably to fix minor errors, e.g. typos, with less effort than provided by WordPress.

   *But using WordPress as an alternative is not a long-term option: WordPress is and remains a lousy platform. Further, the attitude of the WordPress people towards both bloggers and readers is depressing.

6. While virtually all my writings to date have been of a non-fictional nature, I have lately developed far-going plans for a novel.* Regardless of whether this is successful, there will be stretches of time where my other writing and website activities will be correspondingly reduced. It will also likely imply that I prolong my sabbatical considerably.

   *Do not hold your breath: Even in a best case, this will take a long time; especially since I need to develop new skills. Outside of the best case, there is no guarantee that I will manage to complete it and do so with a satisfactory quality for publication.

Excursion on how I tag on WordPress:
I try to pick five* tags with minimal thought spent**. Occasionally, I cannot actually come up with five reasonable tags; somewhat more often, more than five feel relevant. Sometimes I try to pick tags consistent with earlier works; sometimes I try to pick something I have not or only rarely used before; often I just pick the five tags most obvious to me.*** Is this much better than throwing darts? Possibly not…

*Rationale: This is something recommended to me years ago, as a compromise between too-little-too-be-noticed and so-much-that-automatic-mechanisms-think-it-is-spam. Whether this recommendation still holds, I do not know.

**Rationale: This approach of “speed tagging” attempts to make sure that I do not lose too much of any benefit that might be present, while keeping down the time potentially wasted. I am skeptical towards tagging and would rather not tag at all. However, in the days of yore, WordPress had wonderful global lists of posts grouped by tags and sorted by date (that I loved to browse myself). While these grew more user-unfriendly over time and appear to have been abolished entirely years ago, I still cling to the hope that they or some equivalent is still around or will at some point be re-instated. Certainly, some amount of tagging did make sense in the early days of my blogging due to these lists.

***(Ir)rationale: I am torn between a wish to be consistent, a hope to reach someone new in the (possibly imaginary) category listings, and the feeling of just wasting my time with tags.

Advertisement