We use meteor-desktop as well for the reasons mentioned by others already. It is a concern to be so reliant on a package that’s no longer heavily maintained.
On a side note I notice that Microsoft has released support for React Native on Windows and MacOS. With Meteor now providing first class support for react-native, perhaps this is the way forward long term…
Have you thought about using PWA to deliver the app to the desktop? I know it’s not for every use case, but for some use cases it might be good enough.
I’ve been seeing over & over @alawi’s mention of PWAs and somehow I was not aware of their desktop/chrome integration I just made the Meteor Forum a PWA (looks so nice as an app, see image below), not just another tab in my browser anymore.
Looks like the PWA desktop might not give you everything that an Electron app does, but wow, I’m impressed. Not sure how I missed this, but I’m even more excited to make my Vue + Meteor app PWA compatible now!
I had realized this for Mobile, but not Desktop… pretty nice & easier alternatives to Cordova and Electron. Although, I do support keeping strong Cordova support & Electron support growing. I do think many mature apps depend on these.
For me the biggest draw to desktop & mobile apps is just the user having an easy icon to click on and recognize when they use the app daily. This is the BIGGEST factor for users returning daily to interact with an app.
For so many years I’ve wondered why this was not easier to do. Feels like 10 yrs ago now, but I remember using this http://fakeapp.com/ for OSX back in the day, but it was not really a good solution.
It would be great to add PWA details in the Meteor docs. I have noticed Vue’s CLI & Gatsby have clear PWA docs.
Thanks to @jkuester for this Meteor PWA guide, I now finally get why the Meteor core team should prioritize making this more obvious, aka good docs, and maybe an atmosphere package or a skeleton for a PWA app that would configure a package like the Vue CLI does.
@filipenevola hoping you could take a quick look at this and share with the Meteor core team!
@omega a great alternative to getting this into core is getting it into the Meteor-Community-Packages org, so that others can help maintain and release it. Others are willing to help keep it up to date as many ppl have businesses relying on this package. Please see this issue https://github.com/wojtkowiak/meteor-desktop/issues/243
So how does the package get there? As much as I understand it’s no longer supported by the original author.
No responses on bugs/issues either. We went along the path due to security and data privacy reasons and are now facing issues as the automatic update doesn’t work and is leaving our customer being stuck with bugs that are long solved (see Cordova force update on stuck clients due hash mismatch)
We’ve been using electron-updater to update our meteor-desktop electron from electron 2 to electron 9 over the course of a couple years and have never had any issues, so it’s not an issue of meteor-desktop.
If the maintainer abandons an open source project then the path forward is fork it into the Meteor-Community-Packages org. It would just need to use a different npm package name like “meteor-desktop-community” or something
Hi @efrancis and all, sorry, I’m late here, very late btw.
I have never developed an app with Electron but I have a PWA for desktop with Meteor for example.
Electron could also be a nice addition as we need to access restaurant’s printer and other resources in this product (https://www.bemarke.com)
Also, in the Meteor Software (Tiny) company side, we are willing to keep Meteor as a great choice for desktop apps.
What are the issues, PRs that are pending that we could take a second look?
Please send me a private message here or on Slack if you think we can help in any way. It would be important to have people actively development with Meteor and Electron involved in this work.
Yes this is the biggest issue @filipenevola, the package is fantastic and does all the work required to use Electron but the community is unable to contribute and it has a ton of open issues. Bringing it under an official meteor or meteor community org would enable us to keep it moving.
Someone has built a complete integration for Electron and it adheres to Meteors build target concept already, little work is required on Tiny. We just need to be able to maintain and update it.
Perhaps it is time to consider just forking the package? It seems like a good candidate for the meteor community org as there seems to be people willing to pitch in. As it is released under the MIT license there should be no legal restrictions to forking. It’s certainly polite to ask first, but that has already been done.
I second that. The original author hasn’t responded in a long time to both issues as well as to requests to update the package.
Forking it is a short term solution to implement bug fixes that are outlined by the various users but IMO its most important that this package is taken in under the official Meteor community as @efrancis also proposed.
Now that desktop support is being prominently on Meteor.com I think it’d be great to revisit this. For example with HMR in Meteor 2.0 now, getting HMR support for desktop apps would be very helpful for faster development.
Or as another example, releasing an auto updating Electron app requires deploying to an S3 bucket. It would be amazing if Meteor Cloud supported this while deploying so that we could move entirely off of AWS and to Meteor Cloud instead.
@koenlav@copleykj do you still own that meteor-desktop community repo? I’d really like to get this fixed asap and get it published to npm. Mongo Atlas is forcing some tiers to Meteor 5.0 which requires upgrading to Meteor 2.6, and meteor-desktop is currently incompatible with Meteor 2.3+ because of this issue.
That means anyone using meteor-desktop is going to have their app break in production some time next week unless this is fixed and released to npm.