Fork me on GitHub

Is there a benefit over migrating projects to bitbucket or gitlab. Every now and then I see a project in the competitors but not sure what benefits they offer. Bitbucket a while ago mainly was great for private repos on a free account, now github has that. Is it just the "get away from microsoft" draw?


@theeternalpulse I suspect it's mostly just down to whatever service folks were already using for other things.... At work, we use BitBucket because we're an Atlassian shop. When we first started releases things as OSS, we just released them on BitBucket because we already had the code there. We've since switched to using GitHub because it's easier for the community.


TIL there was a browser with haiku error messages. I wish more of the web was like that 😄

👍 8

same as @seancorfield - for work i stick with bitbucket because it plays along well with other atlassian tools


I've only ever used bitbucket because of the unlimited private repos. I remember reading that people liked gitlab for its usability / ui / project lifecycle management, but have not tested it myself


Gitlab with GitlabCI has always been comfortable for me. Also being able host it yourself is cool.

👍 4

Thanks for the input.

Ashley Smith19:07:07

Hey so I was wondering if anyone would lend an ear to something I need help with. I'm going to be in my final year of university next year doing my masters degree in computer science. I really need some help choosing a project as I want to start development asap (writing a thesis is my weakness so I want to plan accordingly). My problem is I truly do not know what I could do. I've decided I want to work with clojure as I've somewhat fallen in love with the language. I want my project to be open-source, as I would love for it to become something I'm known for. I don't want to just make something for uni and abandon it after my degree, I want something that looks good on my portfolio. However, with it being a masters project, I believe that the project itself needs to be something I can test. I am honestly so stumped as this world of academia is still new to me. I don't know if being something I can test is a requirement or not, but I don't honestly know what to do - is a masters project about building something, or experimenting something?

parrot 4
Ashley Smith19:07:15

Once I find the project, my attention span will massively increase. Right now I've just been wandering around academically, as I know that I'd have to drop anything when it's time to start the year anyway. My initial idea was to make a new kind of forum concept based around Q&A and tags. Without going too much into it, the idea in my head excites me as I feel like if I succeed it will either be useful to others, or at least impressive in my own eyes. If I then open source it, people would host their own under their own domains and it'd spread and be fun.

Eric Ervin20:07:01

I've been out of university for 20 years? ouch So I might not know what it is like. I wonder how you'd find someone in your university community who needs something developed and your project would be filling that need. It would be an awesome combination of your programming powers, you partner's knowledge of subject X, and a demonstration of your ability to develop something from someone else's specifications. Maybe that is too much like the world of work 😭 and not like computer science. My graduate school experience was in librarianship and we'd often be tasked with making an archive for organization X, a database for organization Y, a thesaurus for organization Z. That is where I'm getting this mindset.

Ashley Smith20:07:10

I guess it could tie in with my forum like idea, but it's for pretty much any community I guess. I want to simply make something and that be my project, but I need to write research proposal and all that and actually gather data of some sort, 'solve' a problem

Ashley Smith20:07:52

thats where I'm struggling 😞 I need to contribute knowledge rather than just making a product, so unless my project fills a gap like some framework that does x faster or y better, then simply an app doesn't cut it

Eric Ervin20:07:29

Oh well. When everyone gets back to Slack tomorrow, maybe someone who has gone to school in this millennium will have an idea. 👴:skin-tone-4:

Ashley Smith20:07:54

from your photo you look like you're doing fine, relax 🙂

Ashley Smith20:07:22

and thanks for the help. I just wish I didn't feel so lost


hey @ashley - would recommend taking a look at this guy's blog for inspiration


(would also recommend his books, "deep work" and "so good they can't ignore you", but the blog is more student oriented)


This is one of the articles that stuck with me: - he talks about how people found non-obvious opportunities to innovate


"What do I build" is very much a business problem that everyone is facing (what do I build = what will people pay for? what can I get published?) - one idea is to seek opportunities where two specialisations cross. You'll most likely end up writing something to automate a business process, or a monitoring dashboard, or a comparison/analysis tool helping people make decisions.

Ashley Smith20:07:06

that was a really good article but if im honest it just makes me feel even more mundane than I was before 😞


(Another source I would recommend reading is a short book, "The Mom Test", which is on how to evaluate whether your idea is any good and fulfils an existing need.)


Welp. 90% software development is extremely mundane ¯\(ツ)

Ashley Smith20:07:47

im sorry I didn't mean that in a bad way! I guess you could say I have a massive inferiority complex and its kind of bearing down on me now its time for me to do something


That's also common 😄 Don't worry, you'll do fine.


As for what to do for your masters, it depends on what direction you want to grow in. Building something practical gives you a project to show to prospective employers (direction: industry). Building a research project, or some kind of analysis, directs you towards a career in academia.

Ashley Smith20:07:30

I want to do something practical as academia isn't for me I don't think - I don't like the hypothesis, research and conclusion things


Your profile doesn't say where you're studying, but where/when (more than 10 years ago) I was at uni, professors would suggest subjects to students.

Ashley Smith20:07:50

I'd rather just make something that people can enjoy using, playing, working with

Ashley Smith20:07:11

My university isn't that great to the point where I'm quite ashamed to say I'm attending

Ashley Smith20:07:41

They will probably help me when the year starts but its still summer, so can't contact them for another 2 months

Ashley Smith20:07:50

kind of why I wanted to get started, get ahead of the game


Well, you can use the 2 months to... research, then 😉


With a practical project, you have two things to solve:


1. find a subject that will give you a good grade 2. find something that will be useful to others


I would suggest that it's not a good idea to couple those, because finding something that will be useful to others is a hit-and-miss process (even when they assign you a project at work, it may turn out to be cancelled!), and so not necessarily a good way to provide motivation and self-esteem.

3Jane20:07:12 <- this guy is good for articles on motivation and sustaining good habits (also has a book, Atomic Habits)


So maybe use the two months to talk to as many people as you can, and see if they are facing any problems that you'd know how to solve with software.

Ashley Smith20:07:17

you've certainly got awareness of a lot of resources


Several years in a startup (motivates you to think "how do I find a problem I can solve"), and I read a lot 🙂 I will keep my fingers crossed for you. Good luck!

Eric Ervin21:07:42

Can probably get a good amount of programming calisthenics done in those months. What do the kids call it? Tooling up?

Ashley Smith21:07:30

the problem im really having is that I can't think of a way of making a project that both has research with it AND is actually useful

Ashley Smith21:07:38

in the sense that I'd rather design a framework, program or webapp than an experiment, but I think that I have to do some sort of testing regardless so whatever I make has to have something I can measure and analyse

Eric Ervin21:07:24

Is there somewhere you can read all your predecessor's work? Something I remember from college is taking someone else's thing and making it 2% better.

Ashley Smith21:07:21

thats a good idea, I'll try to find osme


Back when I tackled this same issue (final year BSc project, then my PhD topic), I was just fascinated by programming languages themselves so I did projects around language design and implementation because it scratched an itch and kept me interested -- but was also sufficiently research-y to satisfy my supervisors...

seancorfield21:07:26 there something that you're just super curious about as a topic area, that could overlap with Comp Sci @ashley?


(and my PhD project ended up being used to teach FP to undergrads for several years after I left)


You mentioned a Q&A-based forum with tags above -- what would you envisage being different about that, compared to say, StackOverflow? Would the research-y part of it be NLP and identification of related questions and/or automated identification of tagging?

Ashley Smith22:07:42

@seancorfield so the reasoning behind my Q&A based forum was from my own needs. I learn through mimicry, through leeching off of someone who already knows and learning how they think, through talking and asking questions. My idea was to create a forum that goes something like this: You want to implement something in clojure but you don't know how. You ask a question, for instance, 'how do I make this button make a sound when I click it?' and tag it with whatever the current community offers (in a programming community, it'd be languages, in a clojure community it'd be packages and stuff maybe depending on whoevers hosting it). There is a level associated with each question and it starts off with the same level as you (so, this user is new, so has level 1 in clojure). People will either attempt to answer or admit it's slightly difficult through bumping it, thus increasing it's level in some of the tags. If someone tries to answer it they join a thread where they can see previous messages and talk with the user. They can pass and bump up the question's level, or satisfy the user where they will be awarded with some sort of experience like a video game in related tags (so, answering questions levels up your own tags, so when you ask a question it's starting level can be maxed to your own). Obviously there's some areas that need exploring, but the idea is that a lot of new programmers cannot use stack overflow as they are required to produce minimal examples of the errors - but the errors in question may not be obvious to them in the first place. I believe that the outright rejection of questions isn't always the right answer, and so I feel like a way of communicating a little more personally would help newcomers. Users could learn from each other while also gaining experience in their tags. I like the idea of this, as it works with programming but it could work with nearly any topic. Take a video game like World of warcraft or something, the tags could be classes, dungeons or activities, and players could ask questions or even for volunteers to do things. So yeah.. It's not a reasearchy thing, but it is something i feel like I could give a good crack at, learn databases and make a simple prototype with, you know? The only thing getting in my way is that I can't research anything out of this really?

Ashley Smith22:07:47

Sorry that answer took a while to type

Ashley Smith22:07:04

Another ancedote, was that when I was learning how to link SDL with haskell, my error worked on one machine but not another. I couldn't reproduce the error and instead was looking for any problematic code or just anything that could have been OS exclusive etc, but SO refused to accept my question because their site is more for everyone, not the questioner (the questions must ensure that future readers can benefit as well), so this new forum isn't in direct competition and is more of just a way of finding and collaborating easier, you know? My error turned out to be simple - I had forgotten to actually install SDL and was getting mixed up between the meaning of bindings and the actual program the haskell bindings were for, something which any regular haskeller could probably have told me but instead I wasted a night getting frustrated.


So the academic/research-y aspect of this might be gamification of learning through the Q&A process, based on peer rankings... or some such...

Ashley Smith22:07:38

thats a very psychology-y thing though right? Its not like im profiling for efficiency or comparing the output of algorithms which are the kinds of things I'm imagining


You might talk to some profs in other departments about learning processes and see whether there's an "in" to use an exploratory project like this to help their students learn from their peers -- at the same time creating a user base for your project and some "captive" users that could provide feedback.


Right, your part doesn't need to be very psych-y tho' -- you'd just need to track enough user behavior (tagging, pass/bump, answer, reward) to provide that data to the departments that might be interested. In other words, sort of delegate the psych-y part out to the profs in other departments 🙂

Ashley Smith22:07:41

in other words, are you suggesting I 'research' this topic through asking for professors opinions in the gamification of Q&A forums and then simply build the thing with their advice in mind?

Ashley Smith22:07:57

that actually sounds a little bit plausible if I managed to talk to someone


It would also let you focus on the platform-y aspects -- making it generic enough that very different communities could all use it. Yeah, go ask some profs who focus on learning processes and see if they're interested. Sometimes, you can get inspiration from outside the comp sci box...

Ashley Smith22:07:32

does this sound like a masters project or an honours?


Hard for me to say. My BSc final year project was to write an APL interpreter. Then my PhD work started as a Masters and I transferred to PhD after one year -- that was the Design and Implementation of Functional Programming Languages.


I'll be honest, I only did it because a) I was very interested in PL design/impl and b) the department had budget for a PhD project and were happy for me to do pretty much whatever I wanted.


"pretty much" -- my super and I fell out over the amount of academia in my project: he wanted more focus on program proving and transformation, but I was much more interested in expressiveness of language constructs... and garbage collection algorithms (this was the early/mid 80's)

Ashley Smith22:07:58

Sorry I was emailing the one tutor I like so hopefully he can determine if this idea could qualify or not

Ashley Smith22:07:16

Thank you for all of your answers across all these channels by the way 🙂 I like it here!