All Entries (Page 13)

Single-Purpose Computers

Entry created on 2024-07-25 author:steeph (374) cat:#100DaysToOffload (41) cat:Computers (22) cat:Media (1) cat:Music (5) lang:en (253)

A general-purpose computer is a technical term that describes a device by the capabilities its components enable it to make use of in principle. A turing-complete machine. A "smart" washing machine with a touch screen, internet connection and a CPU fast enough to run Doom may be considered a general-purpose computer even though you may only want to use it for one purpose. So, it practice it's a single-purpose device, with a computer. It is this sense in which I have developed a fondness for single-purpose devices. Nowadays, these devices often end up having a computer inside them. But that doesn't mean that it makes sense to either run software other than the one provided for the single purpose on it or replace it with an (already existing) PC/laptop/smartphone/tablet/whatever. I reckon it's fun to hack household devices to run arbitrary code on them, get Linux to run on a calculator or build a botnot of smart fridges to mine bitcoins. But for daily use and the simplicity of it in daily life, I find reliable software that serves a single purpose well, soothing, comforting and simpla practical. Turn on the device, select what it should do, go. I could do without all the complexity that in modern electronics that serve mainly the developers (or their bosses need for a cheap and quick, flexible complex system).

I think many users of modern "smart" household devices would agree. But my preference for single-pupose devices goes further. My laptop and my desktop PC are connected to several home servers and NAS. It's cool to have an MPD server, an always-online, logging IRC client, a LAN-centric torrent client that runs independently from the actual terminal devices that you use. It's practical and clearly has its purposes. But, after working a more than full-time job for years, I've learned to pay tribute to the simplicity, lack of maintenance requirements and reliability of computers that are used to do just one thing. This is different from the Unix-philosophy of doing just one thing and doinf it well. I'm not talking about standardised interfaces between software components. I think that the advantages of a general-purpose media center in the form of a computer have made us overlook the advantages of standalone devices that serve a single purpose. You might have a CD or record player for music, a DVD player for movies, a notebook with a good keyboard for notes or a diary, a thin laptop or tablet for casual web browsing, a Play Station for gaming. Turn it on and it runs. No complicated and long boot process, no chance a software update might interrupt you or have broken something. Hardware buttons for the specific functions the device has. And, if a device breaks, the other ones still work.

I have askey myself why. But I really don't know why I preferr CDs over a music collection on a hard disk. It's just so much more fun to choose an album and put in the disc. Without planning to, I have started to segragate my devices more and more over the last couple of years. I have a PC for retro gaming, one for OS experiments, and so on. Doing those things in VMs is out of the question for me. Again I don't know why. But it really feels much better to have a dedicated device for some things.

Life is just the accumulation of memories.

Entry created on 2024-07-24 author:steeph (374) cat:#100DaysToOffload (41) cat:Life (5) cat:Lucid Dreaming (12) cat:Memories (2) cat:Philosophy (5) lang:en (253)

This is not really a philosophy that I live by or use to form my conception of what my or anybody's life is. But I find the thought that everything that happened in the past is only part of reality in the form of a resedue in our memories interesting. That isn't really true. If I break a cup that I use every day, I cannot use that cup to drink tea from. And if I build a house, I can live in it in the future even though the action of building it and all the experiences I made during that time is nothing but a memories. But applied to activities that don't transform part of the physical reality in a specific and definite way, it has some truth, and realising that gives me a curious feeling.

Somebody asked me once, after I gave an introductory talk about Lucid Dreaming, if what somebody experiences during a dream is somehow more real than a dream or still just a dream. I've answered in a way that I thought as honest but was maybe not a clear answer. I said it's still a dream and what stays after the dream is nothing but the memory of it. That's not wrong. But had I taken the time to give a longer answer, I should have also explained the context in which I consider this to be the case. Regular dreams are usually incredibly volatile, especially for somebody who doesn't pay much attention to them and doesn't even try to remember them better. For most adults that's how they remember the majority of the few dreams that they remember at all. Most lucid dreams, it is often said, are a very positive experience. People who train to become lucid dreams mostly consider a lucid dream a success that comes way less often than they wish. These circumstances alone make a lucid dream easier to remember. Some see the memory of a lucid dream in a different class from non-lucid dream memories all together. But of course there can be lucid dreams that you don't remember for long, or you forget the details after a while, or you don't recall at all after waking up (How would you know?). If you don't consider a dream special, it will fade more quickly. If your head is full of other pressing thoughts, if you're depressed or are currently very worried, and if you don't write them down, the memories of a dream will fade quicker. In my experience, given enough time, the memories of lucid dreams will fade into the same jumble of vague memories from long ago, which might be correct or complete or not at all (which doesn't correlate with the sense of how correct or complete they are, btw). That isn't to say that they weren't worth the effort I've put into.

All of that is also true for waking memories, though, isn't it? Yes, we keep much more of what we experience while awake because the short term memories are functioning a lot better then. But years later, what's left is a fading memory unless it is a special memory to you in some way or you do something to keep the it alive. And the jumble those waking memories fade into is the same where all the dream memories go. So, the more small, unimportantant memories accumulate, the more likely it will become that a dream memory is confused with a waking memory. I believe that, to a certain extent, this may be normal. To an extant to which it is not concerning, I mean. Did I see a deer in the northern fields where you usually keep away from two years ago? Or did I only dream that? Deer have no relevance to my life, nor have those fields or anything that seeing a deer north from the village would imply. So I don't care. This is something else than believing that what you dreamed last night to be true minutes or hours after waking up. That could become an awkward day at work or worse. It is also something else to come to the wrong conclusion from a reality check when you're awake. What I believe to be normal to some extent is the confusion of basically irrelevant memories.

In many ways, dream memories and waking memories are more similar than I thought for a long time. They are the same in some ways. A lot of the apparent differences can be explained by the lack of short term memories while dreaming or during the process of waking up (which can be both at the same time). So, are lucid dreams just dreams of which fading memories are the only thing that remains? Yes, just like your last night out, your holiday in Japan, all the films you've watched and podcasts you've heard. And no, just like everything else you experience, a lucid dream re-shapes and re-inforces neural connections and thus influences how you think, how you experience things from now on and what you will do in the future.

The Bad Web

Entry created on 2024-07-19 author:steeph (374) cat:#100DaysToOffload (41) cat:Computers (22) cat:Open Web (2) cat:The Small Web (3) cat:Web Browsers (5) cat:World Wide Web (10) lang:en (253)

A lot has been written about the declining usablity of the World Wide Web due to web sites not respecting what visitors really want or need. So I'll just summarise here before I'll try to get to the point.

Megabytes of CSS and JS to display 15 Kilobytes of content and another few megabytes of ads and other bloat. It has become completely normal to have a hundred and more tracking cookies installed for wisiting a single web page. Many popular sites can't be read by at least some people because the distracting ads have become too much. These are the first major problems that come to my mind. Depending on who you ask the problematic development has started in the last couple of years, about a decade ago, in the mid-2000s or even in the 1990s. But most people above a necessary age to have experienced the difference seem to agree that the web experience was a better one in 2005 than today. Back then RSS was integrated by many popular sites. You could use it to read Twitter and subscribe to YouTube channels, for example, and sites that still offer it in the background used to place links to feeds visibly, not hidden in the source code for browser extensions to discover them. Web browsers themselves could not only display RSS feeds but placed an icon next to the address bar when a feed for the currently viewed page was available. RSS is often used to show how the web was more open in general. Even commercial web sites were created with a more open approach. A site were you had to register before you could view its content was an exception for which privacy was the reason, not monetary expectation or greed. This is the time to which most people seem to want to return to. When I say most people, I mean most people whose thoughts on the open web I read, which is those who post to the open web and are interested in such things to a degree that they want to write about it. So what I probably mean is "most people who are dissatisfied with the current state of the web". It's possible that most people, or most internet users, love the way things are going now and hate the ideas advocates of the open web have, whether that is for or against their own good.

Sometimes I boost thought-out or new takes on the subject, well formulated demands or promotions of software solutions on the fediverse. And I often think about this myself. Because the web has brought me so many nice things and I want it to be a positive thing in society as well, which, overall, it doesn't seem to be anymore. What does the web need to make it better again?

First of all, the open web isn't gone, nor has it shrunk in size. There way more personal blogs, open networks and non-commercial projects out there than 20 years ago. Even new web forums open all the time. But it's less visible below the very very loud, commercial web. Maybe the greedy web is a good name for what I mean. Not every commercial web site is an example of how the web is devdeloping in a bad direction in my view. I want to be able to get information about a business from the business-owner themselves when I'm interested in their services, for example.

A search engine that returns links to non-commercial sites first, unless you really need information that can only be found on a page of a greedy site. I think- let's just not talk about the many problems (not even just challanges) that such a search engine would introduce if it is to be useful in practice.

A browser that only links to non-bloated/non-tracking/non-greedy/open web sites or warns when a link leads to a less-nice site. Again, I don't have the time right now to list all the problem that there would be if an attempt to implement this would be made. Maybe I'll write another entry about my deeper thoughts on the technological solutions that I mention here. But these thoughts don't contain any real solutions. So I don't know.

Create a literal small web, that only uses resources from and only links to, web sites that are following the same standard (e.g. only (X)HTML4, maybe only CSS2, possibly restriction on JS usage). That is in principle similar to building a whole new network, as is Gemini and Gopher doing. (I know Gopher isn't new, but I reckon the majority of sites is.) I forgot what other protocols with similar aims are there. As far as I know none that are widely used. There are initiatives to restrict the WWW to a smaller or older set of standards. Those probably influence site builders (mainly in personal web sites), but won't change the web as a hole. And so you'll eventually while browsing come across a site that doesn't restrict itself it what it's linking to, or you'll catch yourself linking to a bloated site because it's important to link to the original source of something.

JavaScript needs to be optional again. I've recently come to think that this is actually the one major goal among the technological changes that the current web would need to undergo in order to make it user-friendly and more usable again. In a time where you couldn't 100% expect that visitors were using a client that understood JavaScript, and had it enabled, web developers didn't have much of a choice and built in fallbacks so that a site was still usable without JavaScript. But the number of visiting clients without very good JavaScript became so small that it started to look optional, and in reality became not only optional but even rare, that fallbacks are included. JavaScript really took over the web. I could make so many words around this but don't have much time left this morning. Not only are there sites that are empty without JS loading the entire HTML. Such a thing isn't even special anymore.

If you are creating a new web browser, please include a switch in the GUI that allows to enable/disable JavaScript permanently (until deliberately switched on again) either entirely or for the currently viewed site. Or, maybe make it off by default.

Deutsche Untertitel für A Brony Tale

Entry created on 2024-07-18 author:steeph (374) cat:#100DaysToOffload (41) cat:Bronies (2) cat:Documentary (1) cat:Fandom (1) cat:Films (22) cat:My Little Pony (10) cat:Subtitles (1) lang:de (41)

Es gibt da eine weniger bekannte Dokumentation über Bronies, als die Bronies-Doku. Auch über die bekannte von John de Lancie gäbe es etwas zu schreiben. Aber egal. Einfach vergessen. Die muss nicht gekauft werden. Interessanter (auch weil weniger bekannt) ist die Doku mit dem Titel A Brony Tale.

A Brony Tale von Brent Hodge gibt einen Einblick in das Fandom von My Little Pony: Friendship is Magic indem die Frage "Was sind Bronys" langsam und aus der Sicht von Ashleigh Ball beantwortert, während sie entscheidet, ihre erste Brony-Convention zu besuchen und Fans ihrer Voice-Arbeit trifft. Ashleigh Ball kennt ein Publikum von Fans bisher nur von Konzerten ihrer Band. Als Sprecherin zweier Hauptcharaktere in der Animationsserie (im englischsprachigen Original) wurde sie aber auch unter Bronys ein Name und häufig zu Panel auf Fan-Conventions eingeladen. Es gibt auch Einschätzungen anderer zu dem Fandom, aber keine wirklich kritische Beurteilung oder negative Betrachtungsweise. Ein guter Film für jemanden, den die Frage oder das Thema interessiert. Aber keiner, der weit ins Detail geht oder viel Wissen über die Subkultur vermitteln will. Die Bewertungen haben recht viele Ausreißer nach oben und untern, liegen aber größtenteils im unteren Mittelfeld. Also, nichts besonders Tolles, aber als Teil des Fandoms fand ich ihn sehr gut.

Nachdem der Film veröffentlicht wurde gab es wohl auch mal Pläne, ihn im deutschen TV zu zeigen. Jedenfalls hatte mir Brend Hodge derzeit glaubhaft gemacht, dass es Interesse eines Senders gegeben hätte und quasi nur noch ein Termin gefunden werden musste. Ich hatte damals meistens Zeit und Kopfkapazitäten frei und habe gerne und viel an Gemeinschaftsprojekten mitgearbeitet oder bei Crowd-Sourcing-Aktionen mitgemacht. Mein Angebot, dass ich die deutschen Untertitel für den Film schreibe, um eine Entscheidung für die Ausstrahlung im deutschen Fernsehen wahrscheinlich zu machen, nahm er dankend an. So habe ich gelernt, Untertitel zu schreiben. Praktischerweise gibt es gute freie Software, die die Arbeit erleichtert. Als professionell ist das Ergebnis sicherlich nicht anzusehen. Die Zeilen sind viel zu verbos und bestimmt hier und da zu direkt übersetzt. Aber ich halte das Ergebnis für benutzbar und frei von Fehlern. Aus der Ausstrahlung ist wohl nichts geworden. Vielleicht war es auch ein Missverständnis, weshalb ich danach nie wieder von ihm gehört/gelesen habe.

Aber die Untertiteldatei wollte ich dann doch mal irgendwo veröffentlichen, weil sich das mit brauchbarer Software so gehört. SRT und XML (SubtitleEdit-Datei, nicht TTML) sind im Anhang verlinkt. Der eigentliche Film nicht, weil ich das bestimmt nicht darf. Schade. Naja. Keine Ahnung wie brauchbar die Untertitel heute noch sind, bei den ganzen Tools, die die Untertitel-Arbeit mehr als nur vereinfachen.

File Attachments (1 file)

Alternative Web Browser Engines

Entry created on 2024-07-17 author:steeph (374) cat:#100DaysToOffload (41) cat:Computers (22) cat:World Wide Web (10) lang:en (253)

I don't think I have to spell out the problem with the current shape of the web browser landscape in detail. Almost every HTTP client uses one of the now three big engines, WebKit, Blink and Gecko. Blink, as the big bad Google one is definitely one that nobody should consciously choose to use. Not everybody trusts WebKit much more because it is developed by another powerful global player that inadvertently collects more data in one hand than can be considered healthy, Apply. And Gecko, the one developed by Mozilla has been criticises for being pushed into a direction that is less free and user-friendly than it is expected from Mozilla. Since Mozilla has received large sums in funding from Google for a while it can be argued that any recent fork of the engine has a history of being influenced by Google. All the other large engines are no longer developed and thus not seen as a possible choice for the future. But there are some options to be discussed for users who want to avoid using a browser that relies on one of the big engines. And there are signs that the lack of competitive differences in browser engines will be reduced in the coming decade.

I like that the recent financial development in the SerenityOS/Ladybird project has prompted discussions about alternative browser engines and has shone some light on upcoming new browser engines written from the ground up. The interest in a new kid on the block has been growing over the last couple of years. So it should not surprise that several projects are undergoing and aim to create alternatives to the current big three browser engines. But in my eyes most of them are still not well-known enough. Not all are equaly useful. So I've decided to mention some options that could replace a mainstream browser at least in some use case.

Using Old Browsers

Yes, I'm covering the less practicle options, too. Using an old browser with security issues that won't be fixed not only might add security holes in the system it is used on, it also doesn't do anything to tackle the problem. The development of browser engines that are used in newer releases will not be affected by some people using them less. Choosing an older browser merely sends a signal and possibly changes statistics to look like you now also forget updating for a long time. But there is range of different lightweight browsers with engines that have no trouble with HTML4 and CSS2 and also offer good JavaScript support. KHTML (Konqueror) becomes an option again, Internet Explorer with Trident or EdgeHTML could get another chance and an old Opera with Presto could continue to shine. But you might need to maintain an envirement of outdated libraries to make your choice over a long time.

Flow

The Flow browser with its own HTML engine is developed by ekioh, a company with experience in developing browser for various embedded devices. As a free product, the preview of a Raspberry Pi version is available. But browsers for all major operating systems are planned. It uses an existing JavaScript engine the JS support is better than with other newly written browser engines. But it also doesn't add as much competition in the market in this regard. I have not tested Flow. As a product from a commercial business it didn't look interesting enough, yet, even though it is reportedly relatively mature.

Haphaestus

The Haphaestus TV Browser is a project by Adrian Cochrane that forms a web browser specifically aimed to be easily navigated with with few buttons (like on a TV remote control). It is a very interesting project because it is made up from several individual parts (CSS engine, font rendering, box layout engine, …) that all are written from the ground up in Haskell. A JavaScript engine is not part of the project. The web is nowadays more pleasant to browse without JS anyway. Adrian also recently started a free year-long course that encourages others to write their own HTML rendering engine (surely with the intention to collect experience for future paid programming courses).

Goanna

Of the browsers that use the Goanna engine, I see Pale Moon recommended most often. Goanna is a fork of Gecko that, by now, differes in features from the current gecko engine. It looks like Goanna may be the way to keep using a relatively old engine that supports all majer web standards very well with an actively developed browser that runs reliably in the OS of your choice. It may not be performing as well as current Gecko browsers like LibreWolf, but there's supposedly a smaller chance that their developers have been influenced by payments from Google.

Dillo

Small, realy light-weight, very simple and currently with no intention to pay any attention to scripts. Older versions of Dillo had been ported to many other systems. The current version 3 is only developed for the now big OSs (Linux, BSD, OS X). But it is very light-weight and snappy compared to mainstream browsers. CSS support is still lacking in the eyes of users who are used to every positional property to work. (Float support is also still missing.)

LibWeb, LibJS

The Ladybird web browser is becoming more known now that a company has been formed around it that has been promised major funding for the next few years. The project started as "the web browser of SerenityOS" but has since gained indipendent support and in turn supported hope that a novel web browser with a new engine will be established on the market of web browsers in the near future. The libweb and libjs libraries were started to build an intependent engine behind Ladybird. An application that will be seen as a usable alternative to Firefox in daily use with no need to fall back on another browser is still far away though.

NetSurf

Another small web browser that good HTML, CSS and JS support is NewSurf. LibDOM, LibCSS and Duktape (JavaScript) are combined to create a simple and portable browser. NetSurf can be found on Atari, Haiku, Linux, RISC OS and other systems. I've also found it a few times on my journay through alternative operating systems (about which I still have to write in this blog). It's the portable, small browser.

Servo

This is one that I like a lot personally. A noval web engine written in Rust. There is currently no full-fledged browser application that uses it. But there is a GUI demo that allows to test the engine with any URL. In my experience, web sites with elaborate design that make use of different layout rules and a lot of JavaScript tend to render better with Servo than with a current Ladybird (LibWeb, LibJS). Support for CSS3 rules is also better than with the small browsers NetSurf and Dillo. The potential is more readibly visible with this one compared to other new projects. Maybe the fact that browsers based on Servo only exist for specific devices is responsible for it being less known than Ladybird. No wonder Mozilla sucks up Serve during their project to replace parts of Gecko with re-implementations in Rust. I guess eventually there will be no big difference between the two engines.

Text-based web browsers are probably not an option for most users. Most web sites are designed for a graphic layout only and even when a page is structured well and can be read with a screen reader, sighted people usually prefer a GUI to a text-based interface. But if you want or a use case requires it, text browsers like links2 or lynx are also worth checking out, of course. Another thing that I'd like to mention here because it could be considered an alternative web browser are auditory browsers. But this entry is meant to be about browser engines, not browsers. Otherwise there would be many more projects that I should mention (browsers with a small user base, forks of Firefox, browsers for quick keyboard use).

Edit: Here's an interesting chart about the live of web browser engines since the first one in 1990.