• 0 Posts
  • 93 Comments
Joined 1 year ago
cake
Cake day: July 28th, 2023

help-circle
  • This. Nothing is more difficult than understanding someone’s else code and architecture, and even if you manage that, you’re now stucked with the choices somebody else made and nobody wants that (we want to make our terrible choices!).

    More than a final app, the best thing to publish as FOSS is libraries extracted from it to help other developers build there own products faster. That’s something other may want to maintain when we abandon it. And on top of that, it still help to publish your app using this lib to serve as practical example about how to use your it, of course.



  • I’m using a pi4 8gb as my server, with a pi4 2gb as backup in case the first one dies. It’s a very classic server, running postfix/courier-imap for mails, lighttpd for web, bind9 for dns, ergo for irc, sqlite3 for databases. I also use fail2ban for IDS and cron to run tons of various task. All of that is hosted on a Gentoo linux OS.

    The one thing I don’t want to use is docker. I love docker for development or for deploying the main app at work, but it makes managing updates a nightmare for handling multiple services on my server (most your containers probably contain vulnerable software due to lack of system updates), and it eats resources needlessly. Then again, it’s made possible because I avoid the big webapps that usually need it.



  • “Git hosting” would be more appropriate. Unless that by frontend, you mean specifically web frontend, but that would be weird, because forges also provide the web backend part.

    Sourceforge was the biggest FOSS host in the 2000s, before GitHub (mainly because there was not much centralization to begin with). That train is long gone. :) Sure, the name and website Sourceforge still exist. Myspace, Digg and Yahoo do too. They are basically web ghosts, only an echo of what they once were.


  • Actually, I do use git bare repos for CD too. :) The ROOT/hooks/post-update executable can be anything, which allows to go wild : on my laptop, a push to a bare repos triggers deploy to all the machines needing it (on local or remote networks), by pushing through ssh to other bare repos hosted there, which builds and installs locally, given they all have their own post-update scripts ; all of that thanks to a git push and scripts at the proper paths. I don’t think any forge could do it more conveniently.

    For me the main interest of forges is to publish my code and get it discovered (before GitHub, getting people to find your repos hosted on your blog’s server was a nightmare). Even for the collaboration, I could do with emails. That being said, most people aren’t on top of their inbox, in which mails from family are mixed with work mails and commercial spam in one giant pile of unread items, so it’s a good thing for them we have those issue trackers.






  • Anafroj@sh.itjust.workstoFediverse@lemmy.worldFediverse or Decentralisation?
    link
    fedilink
    English
    arrow-up
    4
    arrow-down
    1
    ·
    edit-2
    1 year ago

    I’m sorry to say that, but you get you’re definition wrong. “decentralized” means “which has no center anymore”. ActivityPub is decentralized. The usual criticism of the Fediverse by peer to peer networks such as Secure Scuttlebutt or Dat is not that ActivityPub is not decentralized, but that it will eventually “recentralize”, like client/server models tend to do, when one instance capture all the traffic (like Gmail with SMTP, we already see signs of that with mastodon.social, but we’re still very far from it to be a center). I think that maybe you’ve been exposed to that argument and misunderstood it?

    What you really want to say is that ActivityPub is not p2p. You can criticize the fact that there is a server/client model behind it, which means that users don’t really own their data and can lost it if the server goes down - that’s a valid criticism.

    To which I would answer that it’s a tradeoff. :) ActivityPub is built on top of HTTP, the well known protocol on which the web is built. This makes it dirt simple to build an ActivityPub app. The difference of adoption rate between SSB, Dat or IPFS and ActivityPub has nothing to do with luck. It’s HTTP and JSON, it’s just simpler (and easier) to build on top of ActivityPub. Not only that, but it’s a w3c standard. Which means, for people like me who have been burnt by building apps on top of the Beaker Browser only to see it abandoned, that we can trust there won’t be any rug pull. That matters.

    And of course, you can also… run your own server (look into self-hosting if you’re interested in that, there’s a vibrant community here on Lemmy about that). If you run your server, then you own your data and the other servers become your peers. The idea that only others (presumably big companies) can have servers is a very centralized way of thinking.




  • They do maintain the simplicity of the line oriented protocol, so I’m fine with that. :)

    That’s the strongest point of IRC, IMO, and why it’s kept so simple : every instruction is a plain text line, period. It makes it incredibly simple to build on top of it. You don’t need to introduce a dependency to a project that probably will be abandonned in a few years, at which point you’ll have to rewrite your codebase to use an other dependency, for a few years. You just open a TCP connection, you read lines from the socket and write lines to it, each line is its own instruction structured in well known fields, and that’s it. It’s so simple!

    As long as IRCv3 sticks to that, they have my blessing. :)



  • Take back the control on your data, that’s the whole point… :) Where are you regularly saving data? Those are the prime candidates. Look at self-hosted alternatives for those services. I know big webapps hosted in docker containers managed by kubernetes is all the rage around here, but you can often find Unix style equivalent for such services, the main advantage of putting it on a server being to be able to access it from multiple devices. But you do you, if you prefer hosting big webapps, that’s fine too. :)


  • I organize my crontab by having group of tasks (the programs, the holidays, the housecleaning, etc). And of those groups, the events (the non recurring tasks) come last. So I just list the crontab (crontab -l) and the list of things to come print to the screen, that block being at the end of the file. It’s hard to do better than a text file to list things. :)

    I don’t know if there is a program that lists like “what is coming this month” if you really want to filter out the rest, but it should be easy enough to write, given the format of cron rules:

    crontab -l | grep '*' | awk '{print $4 "," $3 "," $2 "," $1 " " $0 }' | sort -n | grep -E "^$(date '+%-m')"
    
    • crontab -l : list the crontab
    • grep '*' : keeps only rules (removing blank lines and comments)
    • awk […] : print the whole line ($0), prepend by the 4th field (the month), the 3rd (the day), the 2nd (the hour) and the 1st (the minutes)
    • sort -n : sort everything numerically, so that all tasks are now in their execution date order (I made awk seperate the fields with a , character so it keeps sorting numerically past the first number)
    • date '+%-m : prints the current month, not zero padded (thanks to the ‘-’)
    • grep -E '^date' : keep only lines which starts with the current month number

    You put that in a script (like ~/bin/upcoming_events) and you’re done. And then, you can call it from cron every monday get what’s coming next mailed to you. :)

    This could but refined further to display dates in a more friendly format. But as usual, Unix is your friend. :)


  • I’m going to pass for the crazy person around, but so be it : cron.

    Cron can be easily configured to send mails (MAILTO variable when using standard cron), provided sendmail is available on the system. If a command called by cron outputs anything, it will send a mail with the content, which is useful by itself to warn when something goes wrong with a cron task, but also allows to do things like this:

    0 9 28 9 * echo birthday John
    

    It’s really easy to get used to the syntax, it’s just going from more precise to less precise, so it’s “minute, hour, day, month, *”. The last one can usually be ignored (it’s the day of the week, I must have used it twice in my life). So here, “0 9 28 9”, you read it backward and it gives : September, 28th, 9:00. Piece of cake when you get a bit of practice. And cron is everywhere, so no need to install anything. Although, since I run it on my laptop, I use fcron, which has a nice feature to run ASAP tasks which should have ran if the computer was not shut down. This way, I never miss an alert.

    I use it for recurring notes (like birthday, paperwork, house cleaning tasks, holidays, etc), but also as reminders of specific dates when I expect a delivery, have a meeting, etc. For the most important messages, I make it use a script that will make a destkop notification (with notify-send) and have a voice read the message (with mimic). And of course, I also use it to actually launch programs. :)