musou

musou wrote

i don't have a ton of advice because ultimately i think people just have to be motivated to want to do the things you think are good. but one thing that i have seen work is to try to automate as much of this stuff as possible because then it's more likely that they'll see the benefits without having it feel like a hassle.

for example, i do most of my day job work in elixir, my team's VCS is git, and we have to use jira to track work (i hate jira but that's a whole other story). elixir has some pretty nice tooling compared to languages i've used in the past, so i have a bunch of git hook scripts that run a bunch of stuff on every commit and check for a 0 exit status, and if the script fails then git won't let you commit (unless you pass the -n flag as an escape valve but why would you do that all the time?).

so the code is automatically formatted, then typechecked, then the full set of unit tests gets run, then a static code vulnerability analyzer gets run, on every commit. and then lastly, we have a naming convention for all of our working branches to start their names with the unique designation of a task "story" (ugh) in jira. if the working branch is named according to convention, it will extract and format the jira number and prepend it to the beginning of the commit message. if the branch is named something else, you can manually prepend the number yourself, but otherwise it won't let you commit.

the final piece of the puzzle is 2 other custom scripts. the first installs the aforementioned git hooks, and the second checks to see if the hooks are installed and if not, calls the first script. the second script is configured to run before every test run. this is nice because it means a new developer basically never has to think about installing anything, all the checks just happen automatically. and if the hook scripts won't let you commit, they explain the problem in a way that is hopefully helpful and makes sense.

3

musou wrote

the original F.E.A.R. was really good and not too long. don't even bother with any of the sequels though they took the story in a gross and awful direction and the gameplay wasn't as good either, even though the graphics were shinier

3

musou wrote

i feel like i should add that i don't recommend HostGator not just for technical reasons... but also because their parent company, Endurance International Group, is not trustworthy and they would absolutely screw their customers if it made them $1 more than keeping them.

EIG bought the company i worked for, said they were looking forward to growing our staff when the execs jointly announced it to both companies, then on the morning the purchase was finalized they laid off about 20% of the total workforce including many of my friends. people who worked at the company for 10+ years were crying at their desks as they waited their turn to be called into their boss's office to get individually sacked.

a bunch of the execs at both EIG and my company that arranged the buyout all got indicted for securities fraud for only partially related reasons some months later. they eventually settled the case by paying a measly fine.

it was a shitty time. as i talked with old timers from the other companies EIG bought before us, i came to understand that EIG's entire business model is to do exactly that. they find profitable SAAS companies whose leaders have no more ideas, give them a golden parachute out, gut the workforce, paint blood on the walls, and ride the remaining husk of the company til the wheels fall off, keeping only enough tech staff from the original company to keep the products in maintenance mode.

this is unfortunately a pretty common strategy, i think Yahoo does it even better than EIG does (and there is no ethical consumption under capitalism), but watching that whole thing play out from the inside was a really wild and terrible ride i couldn't get off of for almost a year until i found something else.

all that to say, just know that if you sign up for HostGator, you're signing up for the Flickr of web hosting.

5

musou wrote

i buy my domains from namecheap because they are usually slightly cheaper than others. i would start with getting a domain and then decide what you want to put on it.

if you are looking to make a static site, i would recommend using Github Pages or something similar, because it basically amounts to free static site hosting. that's what teatimer.site uses and it's super nice because i only have to pay for the domain and not for any of the hosting costs.

if you need to have actual backend logic then i would recommend getting a cheap VPS from a place like Linode or DigitalOcean, i have one DigitalOcean box and it's fine if you are comfortable with the linux command line.

i also have a legacy mostly static site on DreamHost that i need to move over to my DO box, i'm just being lazy about it right now. i don't recommend shared hosts like DreamHost or HostGator because they only get you a little bit of functionality over what you could get from a free static site host, but they cost almost as much money as a VPS.

i especially don't recommend HostGator in particular, because i used to work for the company that owns them, and i know just how much their systems resemble ancient and tangled balls of twine.

5