That truck is used at Computex to sell merchandise from.
YMS
I don't understand why this isn't done more often. Publishers announce games early, then have to go great lengths to keep the hype up all the time, then the announced date comes close, but game isn't anywhere near finished, so it has to be delayed, fans are disappointed, developers are stressed. Next date comes close, game still isn't finished, delay it again, fans are disappointed, developers burnt out. Next date arrives, game still isn't finished, but cannot delay again, as fans would really be disappointed now, so buggy mess of a game is released, fans still are disappointed, developers have to work hard to restore the reputation of the game and themselves by repairing the biggest issues, and fans are still disappointed as now things work this way that worked that way before. And anyway it's still not what was promised in the first place.
Instead: Work secretly until the game is in a good state. Release, get good reviews. People get exactly what they expected, as their expectations came from the finished game and not some blown-up early-development marketing visions. Fans are happy, more good reviews.
SteamOS would be a particular poor choice as a desktop operating system compared to basically any other Linux distribution. It uses an immutable file system and reverts all system changes upon every update. That's nice if you don't want to fuck up your handeheld gaming device with some dumb changes, but it's generally not what you will want on a device you use for all kinds of things. Of course, with some effort you can work around this, but then, why don't use a system that doesn't just use such a paradigm in the first place and won't roll back your workaround to make it usable with the next update?
If you have billions of targets to scan, there's generally no need to handle each and every edge case. Just ignoring what you can't understand easily and jumping on to the next target is an absolutely viable strategy. You will never be able to process everything anyway.
Of course, it changes a bit if some of these targets actually make your bot crash. If it happens to often, you will want to harden your bot against it. Then again, if it just happens every now and then, it's still much easier to just restart and continue with the next target.
Von Star hat niemand was geschrieben. Die Titelfrage war doch gerade, warum so eine Nischenperson so erfolgreich (das ist ja messbar) ist. Und wenn du den Artikel liest, der eben absolut nicht voraussetzt, dass du ihn kennst (wenn ihn jeder kennen würde, wäre gerade so ein Artikel ja auch vollkommen witzlos), hast du schon deinen Horizont erweitert.
Natürlich deutlich geringer? Was wiegt dein Kanister leer denn so? Google liefert mir für 10-Liter-Plastikkanister recht unterschiedliche Gewichtsangaben, von 350 g bis 750 g ist alles dabei. Tendenziell sind die schwereren wohl für die Wiederverwendung gedacht (Outdoor-Wasserkanister etc.) und solche Einmal-Verpackungen daher eher bei den leichteren zu finden, aber 40 Standard-Shampooflaschen zu 250 ml und 15 g (ebenfalls ein schnell ergoogelter Wert) lägen mit 600 g dennoch recht bequem im selben Bereich wie die Kanister. Größere Gebinde muss man halt auch entsprechend stabiler bauen, mit mehr Materialeinsatz. Also: Absolut vorstellbar (und von den hier schnell vorgebrachten Zahlen her ja auch durchaus möglich und plausibel), dass man mit dem 10-Liter-Kanister tatsächlich Plastik spart, aber für eine natürlicherweise deutliche Ersparnis würde ich dann doch erst mit den entsprechenden Zahlen annehmen.