Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Bespoke software, and a really simple RSS aggregator

routley.io

Post reply on HN
Print view View original

1–10 of 12 posts

Re: Bespoke software, and a really simple RSS aggregator

#3

Post by petercooper »

Oh neat! Coincidental timing for me too seeing this as I'm procrastinating finishing up my own similar thing but built on totally different infrastructure (Ruby, Lambda, DynamoDB, S3, so far..) and for a different set of feeds (several hundred tech engineering blogs). GitHub Actions is something I did consider, but I'm trying to go AWS for everything recently to learn the ecosystem more. (The initial results of my own experiment are at http://engblogs.s3-eu-west-1.amazonaws.com/index.html but it's going to become a 'proper' site as part of the next step.)

Re: Bespoke software, and a really simple RSS aggregator

#7

Post by icebraining »

I have to admit (with a bit of shame, as a Free Software fan) that most of the software I write on a personal basis is like this: bespoke & unpublished. It's just too damn work for what I expect to be a pretty small audience.

For example, I have a small cron script that emails me today's showtimes for my local film museum (which shows old films on a large screen) along with the predicted score from Criticker.

It would be cool if there was some very easy way (I'm lazy!) of letting a non-techie benefit from this, without me having to deal with a publicly available server, storing their Criticker credentials, etc.

I guess the easiest way would be a Chrome extension, but I'm not a fan of the whole app store part, plus I don't want to have to rewrite in JS. And that still leaves mobile users in the cold, which is dumb.

Re: Bespoke software, and a really simple RSS aggregator

#8

Post by ssivark »

icebraining wrote: post #7

I have to admit (with a bit of shame, as a Free Software fan) that most of the software I write on a personal basis is like this: bespoke & unpublished. It's just too damn work for what I expect to be a pretty small audience. For example, I have a small cron script that emails me today's showtimes for my local film museum (which shows old films on a large screen) along with the predicted score from Criticker. It woul…

Basically, we’d like everyone to run their own server where they could add your process as a unit, rather than you running a server for everyone.

I guess this might have been very useful as a Huginn module. Maybe Urbit matures to be a functional version of this. The problem is that today’s computing culture grooms people to be dependent on others’ servers/services, rather than being in control of one machine of theirs.

Re: Bespoke software, and a really simple RSS aggregator

#9

Post by nighthawk454 »

icebraining wrote: post #7

I have to admit (with a bit of shame, as a Free Software fan) that most of the software I write on a personal basis is like this: bespoke & unpublished. It's just too damn work for what I expect to be a pretty small audience. For example, I have a small cron script that emails me today's showtimes for my local film museum (which shows old films on a large screen) along with the predicted score from Criticker. It woul…

I often have a similar problem. It's hard to get around server requirements in a simple and cheap way. And very hard to do simple cross-platform without JS.

Barring that, Github Pages is pretty good. You can get pretty far with a webapp that has some degree of caching/cookie/offline storage. Probably enough to display your local showtimes and maybe even interface with the user's Criticker – presuming Criticker has a CORS-enabled API. Automatic cross-platform UI and mobile support, no app stores or extensions. With Github Actions, I think you could even webpack a React app if you wanted.

But running it on a schedule and sending an email.... suddenly much more work.

Post reply on HN