Ask HN: Which project does not have any good open-source alternatives?

Open Source is changing the tech industry. But there are still a lot of space where the closed source proprietary software is the only player.

Can you comment some of the space? So a person who is thinking about creating a new open source project can get the idea and start making a alternative?

It is about existing open source projects. It would be better to have some idea about new open source projects that are not started yet but needed by the community.

There are some projects that have solutions, but not great ones. My definition of great is something that people with very little technical knowledge can set up themselves, correctly and securely. This is based on my experience, trying to show non technical people how to self host some things.

- File, email self hosted and e2e encryption has Nextcloud, but only technical people can set it up. There should be a happy-clicky thing that non technical people could set up and it would force them to do it the right way, because the right way is the easy way. By happy-clicky, I mean, they create an account on any VPS provider, then click "add service", say how big to let it scale to, define what domain name to use and they are done.

- Web based chat servers that scale. Again, there are web front-ends, like Convos and TheLounge you can put in front of IRCD, but non technical people don't have the patience to set up IRC networks and put web front-ends on them. If there was a happy-clicky solution that used RPC over HTTPS or something to negotiate clusters and all the person has to do is point DNS to it, then more people would self host chat.

- File sharing. If there was something lighter weight than NextCloud that could auto-scale in any VPS provider and could be deployed from each VPS provider, then more people would self host. I almost wrote something like this out of necessity but so many people were using dropbox, I abandoned it.

In summary, I think the gap that needs to be filled is to have more projects that integrate into VPS providers (by the VPS provider) and be part of their VM deployment system, so that anyone can deploy them easily. Maybe there could even be a business model around integrating 3rd party services into each of the VPS providers, following some non vendor specific standard, that all the VPS providers could set up. Linode, Vultr, AWS, Azure, Softlayer, OVH, Hetzner, Digital Ocean, etc... and no code specific to any of them. All of them should have a standard API that services can be integrated into and then easily deployed securely and correctly by less than technical people.

Excel alternatives might be uncountable. Implementing spreadsheet basics is an advanced beginner exercise. But even Google’s billions only get it a distant second best because Microsoft is still working hard despite the lead. Sure Google and Apple can meet most needs most of the time. They’re good enough mainly because they are free beer. Not because they are open source. Obviously.

Don’t get me wrong. I’m a GNU fanboy at heart. But I dont believe in magically willing well engineered software into existence. It takes resources fungible with money. Projects with cash are at an advantage. Projects with cash, motivation and years of experience in the codebase are the ones that dominate. Not starting from scratch becomes a huge advantage.

A new project in a space needs to find an unserved market segment with resources fungible with money. That’s why successful developers’ tools are easy and CAD programs hard.

If you want to make an alternative, just make it and learn from what happens. Because what everyone wants is a free iPhone. Good luck.

That is really True. The APM world does not have any good open source alternative, Elastic APM is ok, but I think it is open core, not open source. Sentry is trying to build one, but maybe that will not be also OSI approved license.