initial commit

blog/FTL.md

@@ -0,0 +1,44 @@
+---
+title: FTL tips
+description: We are not responsible if these tips cause your ship to implode/explode. Contains spoilers in vast quantities.
+created: 16/08/2017
+updated: 08/03/2020
+---
+* Use Cloaking *after* enemy weapons have fired so that they miss.
+* Upgrade Shields as soon as possible; many early-sector enemies can't get through 2 shields.
+* Any of your crew inside the medbay will be healed, they don't have to be standing still. With a lot of pausing and fiddling with crew movement you can heal 4 crew in a 3-slot medbay this way.
+* Upgrade your Doors early on if your ship has trouble dealing with boarders. Higher-level doors slow down boarders trying to break through them.
+* 2 Zoltans in a shield room leaves one unionizable shield layer - Zoltan-provided power ignores ion damage.
+* Upgraded sensors can show you where crew are and even weapon charge levels, and thus may be worth considering.
+* The enemy AI appears to consider Shields the most important system and will prioritize this over other many things.
+* With the Rock C, naming the Crystal crewmember you start with "Ruwen" gives you a quest marker in the Rock Homeworlds. This leads to the Crystal sector portal.
+* Upgrade your engines!
+* You can teleport mind controlled enemy crew onto your own ship, and then kill them.
+* Dropping your shields before an ion blast hits might make it hit somewhere unimportant instead of removing a shield layer for >=5 seconds. They do take some time to come up again, though.
+* Killing the crew of a ship - usually by boarding, but there are many other ways - gets you more loot than destroying it.
+* A Defense Drone I can often be better than Cloaking - assuming you have Shields.
+* Firing two missiles can overwhelm a single defense drone. They also try and shoot asteroids.
+* To unlock the Crystal Cruiser:
+    1. Get the Damaged Statis Pod from a Dense Asteroid Field.
+    2. Find some Zoltan scientists asking to take readings, and get them to open it.
+    3. Go to the Rock Homeworlds.
+    4. Find the quest marker and activate the portal.
+    5. Find the quest marker in the Crystal sector. Then you get the ship, along with some loot.
+    6. Brag to all your friends, shouting things like "HAHA, I unlocked the Crystal Cruiser!"
+* Most stuff doesn't need to be on all the time - oxygen can stay off for a while without *horrible* problems, you barely ever need your medbay and engines/shields only need to be on just before a volley (though shields significantly before). Basically, be good at power management and pause lots to optimize this.
+* Upgrade your engines more! Also subsystems and stuff like oxygen; the upgrades can often be unexpectedly helpful.
+* Don't constantly stay 100% repaired (from shops) as free/cheap repairs come along frequently.
+* Hostile sectors provide more loot too, as there is more combat. This also, obviously, makes them more dangerous.
+* If a ship jumps away, you can jump before it does and skip the penalties (they often lead the fleet after you, etc).
+* Beam weapons cannot miss, and can also go through shields if they have one or more damage per room than the ship being targeted has shield layers.
+* Pausing to think or plan, or really just all the time, is a good idea.
+* If a ship is incapable of actually damaging you, you can use it to train your shield/engine/pilot operators (sometimes weapons) without manual interaction to make them more effective.
+* It's a good idea to save some scrap for emergencies or random good things you may encounter instead of spending it all at stores.
+* Visit as many beacons in a sector as you can without the fleet catching you.
+* Fire weapons in volleys so that stronger ones can hit the enemy ship after weaker ones take down shields. Autofire is bad for anything but ion weapons.
+* Buying crew at stores is generally not worth it, unless you are really low somehow.
+* Do not waste missiles. But also do *use* them, as missiles are often cheaper than the damage you might sustain if you avoid them too much.
+* If your crew are not good fighters, you can open the doors to rooms containing boarders to vent out the oxygen, so they will slowly die. They can break the doors, though, or go through them freely if your doors are only level 1. If the situation is very dire and you aren't in ship-to-ship combat it may be worth moving all your crew to the medbay (or rooms adjacent to it, if there's not space in there) and venting all the ship except that. Of course, if the boarders are damaging your oxygen or door system (or just close to it), this is a bad idea and you should not do it.
+* Don't jump to stores if you don't have things to sell or lots of scrap. You're missing out on scrap you might otherwise get (this is called "opportunity cost").
+* Don't expect to win all the time or even particularly often. FTL is a roguelike. It's designed to be played repeatedly, not won all of the time, and involves heavy randomness.
+* Ships' starting weapons are actually generally pretty good, often using less power and/or firing faster than purchaseable alternatives.

blog/new-website.md

@@ -0,0 +1,9 @@
+---
+title: New site design!
+created: 25/01/2020
+---
+If you visit this frequently (why would you?) you may have noticed that since late last year there's been a page notifying you of service disruption and no actual content any more.
+This is because the old code used to generate the HTML files making up the site was a several-year-old and somewhat flaky Haskell program, and I've been meaning to replace it for a while, but finally got round to it recently after quite a while of just leaving the service disruption page up.
+
+This new site design should be somewhat easier for me to work on, has some extra features I wanted, comes with a nice new style, and should still allow most of the features you're used to to work.
+The service worker for caching is much better (actually works properly now), so you can also use it offline to some extent!

blog/not-everyone-must-code.md

@@ -0,0 +1,39 @@
+---
+title: Not Everyone Must "Code"
+description: Why I think that government programs telling everyone to "code" are pointless.
+slug: nemc
+updated: 09/02/2020
+created: 16/08/2017
+---
+
+Imagine that some politician said "Cars are an important part of our modern economy. Many jobs involve cars. Therefore, all students must learn how to build cars; we will be adding it to the curriculum."
+This does (to me, anyway) seem pretty ridiculous on its own.
+*Now* imagine that the students are being taught to make small car models out of card, whilst being told that this is actually how it's always done (this sort of falls apart, since with the car thing it's easy for the average person to tell; most people can't really distinguish these things with programming).
+This is what the push to "learn to code" in schools (this was written in 2017, when the UK's government was making a big thing of this, but it still seems to be going on now, in 2020) seems like to me.
+In fact, it's even worse, with hyperbolic claims of it being "the new literacy", often made by people who have never done anything beyond basic block-dragging in Scratch or some equivalent.
+
+The average person is definitely going to have lots of interaction with things which have been programmed by someone else, given the increasing popularity of mobile phones.
+This does not, however, mean, that they must know every detail of how they work (not that, at this point, anyone *can* - they're just too complex), and they wouldn't actually be taught this by the "learn to code" things now done in schools.
+Most of the "learn to code" resources, especially those done in schools, start with very simple, visual, 2D-graphical environments.
+This is fine for learning a few basic things (though not very good - Scratch's weird programming environment maps poorly onto actual widely-used languages), but there doesn't seem to be anything beyond that taught most of the time - it's considered "too hard" for the students involved, usually, or there just isn't anyone qualified to teach it.
+And that basic dragging around of blocks is not hugely useful - it doesn't teach much (*maybe* basic concepts of flow control), and you may have to *un*learn things when moving to actual programming.
+
+I have an alternative list of things to teach which I think might actually be relevant and helpful to people in a way that making a cat dance on screen by blindly following a tutorial is not:
+
+* an introduction to computer hardware (for troubleshooting, etc) and what all the myriad cables do
+* basics of networking (what routers do, ISPs and their job, one of those layered network models, HTTP(S), DNS)
+* privacy in the digital age (i.e. maybe stop giving Facebook/Google/Amazon all your private information)
+* operating systems, what various programs are for, and the fact that ones which *aren't Windows* exist
+* what programming involes
+* basic shell-or-equivalent scripting (though this may not actually be very useful either, as the OSes people mostly interact with now - iOS, Windows, Android, etc. - disallow this sort of thing or don't make it very useful, sadly)
+* fixing basic problems using advanced IT techniques such as "using a search engine to look up your issue" and "blindly tweaking settings until it does something"
+
+This doesn't really sound as fancy as teaching "the new literacy", but it seems like a better place to start for helping people be able to interact with modern computer systems.
+
+## Update (09/02/2020 CE)
+
+Having shown someone this post, they've suggested to me that Scratch is more about teaching some level of computational thinking-type skills - learning how to express intentions in a structured way and being precise/specific - than actually teaching *programming*, regardless of how it's marketed.
+This does seem pretty sensible, actually.
+I can agree that it is probably useful for this, since most people will enjoy making visual things with direct feedback than writing a bunch of code to print "Hello, World!" or sort a list or something.
+Still, it definitely does have limits for this given that it's quite lacking in control flow capability and abstraction compared to regular programming languages.
+Also, it's not really marketed this way, and thus probably not taught that way either.

blog/on-phones.md

@@ -0,0 +1,16 @@
+---
+title: On Phones
+description: My (probably unpopular in general but... actually likely fairly popular amongst this site's intended audience) opinions on smartphones today.
+slug: phones
+created: 16/08/2017
+updated: 24/01/2020
+---
+* Why notches? WHY? Just because Apple used them doesn't mean every single manufacturer needs to start making their screens have ugly black bits on them.
+* Really, why Apple at all? They sell uncustomisable and locked-down stuff at higher and higher prices with fewer and fewer nice-to-haves (e.g. headphone jacks) each year. And the vendor lock-in (Lightning headphones, iVersionsOfMostOtherSoftware) is also bad.
+* It would be nice to *not* have battery life ruined to get slightly slimmer phones.
+* The complete lack of updates after a year or so is annoying. Custom ROMs kind of fix this but many aren't available (or even aren't possible due to locked bootloaders) on many devices. This is probably just planned obsolecence.
+* Most of the stuff manufacturers preload (their own UI skinning, apps) is just useless bloat. Especially the preloaded, unremovable, probably-spying-on-you Facebook apps which are annoyingly common.
+* The lack of SD card slots is, again, probably just planned obsolecence.
+* Proper physical QWERTY keyboards would be nice, though as they're such a niche feature that's probably never going to happen except on a few phones.
+* The screens don't need to get bigger. People's hands aren't growing every year. And they don't need more pixels to drain increasingly large amounts of power.
+* Removable batteries should come back. When I initially wrote this in 2017 or so, they were pretty common, but now barely any new devices let you *swap the battery*, despite lithium-ion batteries degrading within a few years of heavy use. I know you can't economically do highly modular design in a phone, but this is not a complex, technically difficult or expensive thing to want.