SoylentNews

Aha! It's been a year since I cast some thoughts to the wind here. There have been countless political panics, social spasms, strange stories, disorderly digressions, and ornery obscurities since. How about we pull up a chair or lift our newfangled standing desks and talk about some of the things that brought us here, whether by way of a certain difficult to pronounce, two character site or otherwise?

I happen to have gotten a turbo XT recently. Besides an immediate affinity for the software that runs on that machine, I've found that 8086/8 assembly is an unusually refreshing reminder of what personal computers both are and can be. The most difficult part of the process isn't even writing the assembly itself! It's in finding useful nuggets of information in the web development fueled muck that represents current technical search queries.

It's been fascinating to learn first-hand about why certain parts of DOS behave the way they do, from the 8088's strange division behavior leading to those comma separated file size listings to the quirks behind memory management and beyond.

Given the above, would you, dear reader, happen to have any old references on hand from decades past? Any fun stories about these 8 and 16 bit machines or their contemporaries? Any old grudges you'd like to air about the latter?

As I sit here with an ISP unable to resolve PPPoE authentication throughout the city in which I live, I find it important to remember how grossly arrogant, incompetent, and useless big tech companies actually are after you peel back their veneer. Google, for example, arbitrarily decided to completely break an email account of mine with no recourse -- whichever chromosomally gifted individual on their team who designed the service never thought to consider that some people may deem functionality (and privacy! How dare I never share a phone number with them!) to be more important than whatever trumped up definition of security they have.

It's gone, vanished, kaput. An account consistently used for five years disappeared at the stroke of some mental invalid's whim. I'll never get it back because the brilliant minds at Google are incapable of determining if it's even my account. It must be a shame being so pitifully overpaid. Someone with more than a single neuron in use could surely use the money. After all, we used to have a very simple and effective system to make such determinations. I believe it was known as password authentication.

Such is life. I applaud you for attending my rant and hope you don't make the same mistakes I did.

The title says it all, really. They have managed to stay completely focused and on topic for slightly over a decade now. While Firefox has gone through about 4 major UI redesigns in the same period (some of which were of questionable merit), SeaMonkey has integrated backend changes without disrupting its users' workflows.

Their project has more functionality than both Firefox and Thunderbird in a smaller footprint with a fraction of the support or market share. As such, it's a shining example of what happens when developers don't succumb to the urge to endlessly churn through minor additions or removals in the name of justifying their existence.

"All the extreme speed and snappiness of Vista combined with the aesthetically pleasing UI of 8"

I have had the misfortune of dealing with Microsoft's latest and worst OS on four different machines now, ranging from the low end to the upper mid range. In every instance, I have found it to be a painfully slow trainwreck of an operating system that even I, as someone who actually likes Windows Vista, find painful to use.

Are you planning on installing it on anything slower than an SSD? Good luck. In one instance, it took well over 20 seconds just to get the file explorer running from an empty desktop.

Do you like having decent graphical effects and a consistent user interface? Have fun with that then. Window controls are now completely monochrome and icons seem to have been designed by someone with only MS-Paint, a monochrome display, and an understanding of BeOS at their disposal.

Do you like having a computer that is responsive and does what you want? Windows 10 is not the OS for you. Beyond the problems listed before, you get to enjoy forced updates on a schedule that suits only Microsoft. Instead of the 30 seconds often possible with Windows 7 and 8, you can enjoy tens of minutes of updates at random on any startup or shutdown. It's Russian roulette for your productivity!

For the past decade or so, I have been accumulating a large number of older video games, consoles, and computers from a large number of manufacturers. It has recently been brought to my attention to start checking online to see what all of this stuff is worth.

My response is as follows: I sincerely hope that the kids who grew up in the '90s figure out how to manage their goddamn money. I understand that many of them are feeling pangs of nostalgia for when they were younger, but they really need to stop putting up these prices.

Let's consider a few examples:
1. A Sega Genesis model 2 with two controllers is generally priced around $30 on eBay. This seems fairly reasonable, as the Game Boy Advance and PS2 cost about as much.

The SNES, on the other hand, costs three times as much. Which console is better, of course, is debatable. The only problem is that both were sold in roughly equivalent numbers, had similar mind-share, and were home to a large number of outstanding games. As such, I consider the SNES to be absurdly overpriced for what it is (especially since you can now buy outstanding clone consoles and digital copies of the games for the platform at a significantly lower cost).

2. The Nintendo 64. Well, not so much the console itself (which can be had with a few games and controllers for around $75, which is good relative to the SNES) but its games.

The obligatory Mario game? $30.
Ocarina of Time? $35
Smash Bros? $55+

Its competitor, the PlayStation, can be had for less than $30 with two controllers and 6 or more games.

Before anyone argues that the Nintendo 64 is worth more because of how poorly it sold, it should be pointed out that you can get a Sega Saturn, the undisputed loser of its generation in those terms, for $90 with 8 games and two controllers. Furthermore, the Nintendo 64 had one of the smallest libraries of good games in recent history.

Some background:
My first Linux distro was Yellow Dog, which ran KDE 3. I found that I really enjoyed the flexibility that KDE had to offer and saw that it was far nicer to use than GNOME, which was also installed. Fast forward a few years, and I find that KDE4 is simultaneously visually stunning and just as flexible as KDE3 was. I like KDE4; I wouldn't be using a compositing desktop environment if I didn't enjoy having nice visual effects. Things like depth, gradients, transparency, and actually being able to inform the file manager that you prefer double-clicking are all very appealing to me.

This week, I decided to try the latest release of Kubuntu on a workstation I was supposed to set up for myself. Unfortunately, as I would learn, the latest release of Kubuntu has KDE5. My first thought when looking at the flat, pastel mess that lay before me was, "Hey, this is KDE! I can fix this." Then I found that the developers disabled transparency in window borders and effectively crippled custom styles. I proceeded to try to at least set the file manager to my preferences. The option to double-click was either moved to a rather difficult to find location or removed entirely.

Frustrated and frankly disappointed in the developers for falling into the same "the users be damned" trap as Gnome, I resolved to just build KDE4, since choice is clearly the strong point of free software.

Or at least I would have, if the developers hadn't been purging the documentation for KDE4. (https://techbase.kde.org/Getting_Started/Build/Distributions/Debian is one example)

Or if they hadn't resorted to a complete clusterfuck build system. (CMake, by itself, is great. The way that they chose to segment the project, distribute it, and handle CMake in context is the clusterfuck).

Or if someone had cared enough to either archive the project or fork it as was done with the Trinity Desktop Environment.

I just found out about Oracle's current generation of SPARC hardware and was wondering how much it cost. An entry level T7-1 server is about $40,000, and I couldn't stop wondering why they have such a terrible strategy with these machines.

I'm not saying that the hardware isn't absurdly difficult to manufacture or that they aren't selling "enterprise" products. What I do believe is that they really should be selling cut down or defective versions of their current SPARC chips for the workstation and small server market. I would fully expect that they are throwing away a very large number of chips that could be better used to grow their ecosystem and attract developers. That way, they wouldn't have to extort a dwindling number of interested buyers for their hardware and could have a stable income source for the future.

According to the National Bureau of FUD, 189% of all tasks that can be done on personal computers are undeniably better suited for mobile devices, especially such tasks as video rendering, content creation, high performance computing, data entry, industrial control, gaming, and mass data storage. In light of these shocking statistics, we must ask ourselves whether the Personal Computer is dead? Then ask ourselves the same question again. And again. And again. And again. And again. And again.

        Why ask the question so much? Because a group of computer users, who are obviously stuck in the past, keeps on insisting that the personal computer, what with its disturbing open standards and accountability and user-serviceable parts could possibly provide some sort of equivalent experience to utter glory, nay, utter perfection in the form of mobile devices. According to reputable sources with whom I have made acquaintance, touch interfaces are vastly superior to traditional interfaces such as keyboards and mice in terms of speed, durability, smudge-capacity, reliability, and accuracy. It is clear that the market has spoken on these matters, as, according to figures provided by the Microsoft Institute of Making Windows 8 and 10 Sound Like Good Ideas, mobile devices outsell personal computers by a factor of one hundred million billion to one.

        Fortunately, there are ways of assisting users of personal computers with their apparent mental illness. There is a very large and vocal contingent of PC software developers who are ready and willing to make mobile-like user interfaces across every single modern application, regardless of whether it “makes sense” or “actively impedes users”. Things like sandwich menus, over-sized buttons, endless hidden levels of displays to change even the simplest of options, forced updates, no promises of stability, the removal of clearly labeled icons, graphics drawn by a mentally deranged and disabled seven-year-old, the destruction of operating system choice, endless tracking, and large swathes of bright pastel colors are sure to cause even the most hardened “power user” to come to his or her senses and embrace the superiority of mobile platforms.

        Microsoft, to their credit, has led the charge on ensuring that personal computer users can experience the joys of mobile computing, even with mice, keyboards, and all of the negative features mentioned before. Microsoft pioneered the mobile-ification of user interfaces with Windows 8, a monumental, immensely popular, and well received step forward for the company. Just three years later, they even had the grace to not only make Windows 10 upgrades free, but to force users to upgrade whether they wanted to or not (this was done to take care of the last few holdouts stranded with “well designed” desktop operating systems). Furthermore, because Windows 10 mandates Secure Boot, users can feel safe in the knowledge that no one can ever change their operating system to some inferior, non-mobile version and that, in order to upgrade, they can experience the unparalleled joys of buying an entire new computer!

        Even open source software developers are taking up the challenge of converting their existing user bases. Just look at the unparalleled successes of Gnome 3 and Ubuntu's Unity user interface! Not only were they well received, but they caused a number of people to become converts to the One True Platform.

        In conclusion, the personal computer is dead. There is no question about it. Anything that personal computers can allegedly do is done better and faster by mobile devices, regardless of what “users” say.

I think that it's time to build a new distribution of Linux, one focused on getting the job done. I think that most distros have basically lost the point of building a linux distro: to have your own unique ecosystem which provides for a specific set of needs.

I'm naming my small, hobby-level efforts to build a distro "nu", as it both provides a very nice logo to work with and because it as a symbol is used to represent "Degrees of freedom" in statistics.

Since this is effectively just a hobby-level of effort, don't expect big things of my distro (unlike a certain Debian fork), but if it works, it should provide a fairly decent starting base for the development of a solid server and development OS.

Some intended features of nu are:
1. nupack, a package manager that effectively just wraps shell scripts and data in a gzip'd tarball.
2. nuinit, an experimental (read: based on random speculation/ideas that I have) that uses makefiles to start and stop the system. If this works properly, it should provide all of the benefits of a certain invasive init system (ie. multithreaded execution) and a model for managing dependencies that is already tried and tested. If this doesn't pan out, I'll likely search for another init system starting with Epoch.
3. nubus. I understand that this was the name of a system bus from the late '80s and early '90s, though I'll likely come up with a better name later. This would start as a re-badged dbus with future revisions actually being their own thing.
4. Either the alsa or jack sound systems.

Because the primary aim of this project is to make a serious operating system, I don't intend to include bloat such as KDE or GNOME. The user is, of course, free to build and install these things for themselves, but they shouldn't expect me to support it.

Project status:
I'm currently trawling through Linux From Scratch, which will provide me with a base system to do various experiments and development on. Since LFS is a very small and basic system in and of itself, it will make it far easier to test out various init systems and implement my package manager.

Future work:
I'm going to implement the package manager first, since it should be fundamentally "easier" than the rest. Thankfully my computer can do the LFS Standard Build Unit in about a minute, so builds have been very fast and straightforward.

A recent article posted here reminded me of all of the great FOSS games that I've gotten to play over the years.

To be honest, while FOSS games do feel kind of rough and unfinished sometimes (though it really does depend on the game; BosWars and LinCity feel like very well made games), they tend to be on a level of stimulation and creativity that really hasn't been seen since the 1990s.

1. LinCity (as well as LinCity-ng)
LinCity, while likely to draw a direct comparison to a similarly titled game by Maxis, really is in a league of its own. LinCity is a construction and resource management simulator that actually leads its players to think about how to design and build a city at a far more involved level than traditional city planning games.

2. BosWars
BosWars is an awesome RTS game that is absurdly easy to modify and improve (the game engine interprets plaintext scripts written in LUA). While the AI either lands in the categories of "Lacking" or "absurdly hard", I could see that it would be an awesome multiplayer game.

3. Globulation
Yet another RTS game, except that this one puts you in a far more interesting role. All the player can do is set construction and general goals/waypoints for the glob creatures presented. From there, they automatically go about their work and assign themselves to different buildings and waypoints. I think that this game would also be really awesome in multiplayer, though the AI is geared very well for every level of skill ranging from beginners to advanced players.

4. OpenTTD
OpenTTD is a free and open source clone of Transport Tycoon Deluxe. As such, it inherits its parent game's features while adding a wide variety of tweaks and improvements, such as a built-in modding utility/package manager.

5. OpenArena
This is a fork of the classic Quake III Arena, and thus inherits that game's fast paced action. While I'm not really too much of a fan of FPS games, this one is really fun to play. Please note that the character models used have quite a few polygons, so it is worthwhile to roughly double the system requirements listed on their page for a playable experience (though I would expect someone with SLI'd Voodoo 2 graphics cards to do pretty well on the processor listed).

6. KSpaceDuel
This is a clone of Space Duel. Nothing more need be said except that it has quite a bit of customization options available.

These are just the games that I have personally played and enjoyed. Wikipedia has a far larger list, and I encourage anyone reading this journal to check it out and reply if they so wish:

http://en.wikipedia.org/wiki/List_of_open-source_video_games