@benstr and @sashko … seriously enjoying the podcasts! @sashko - your explanations of the MDG thought process about various topics has been illuminating. It was great to hear your perspective about what happened with Velocity.
In that same vein, I am wondering if you gents would consider a discussion of the aging version of node in meteor. What are some of the technical challenges with updating to the latest version? I hear some people say that Fibers is holding it up, but others say that isn’t it. What is the real situation with it? What are some of the benefits of the latest version of node vs v10.4 (i think) running in Meteor now. And is a node upgrade for Meteor on MDG’s mind?
This was - I think - actually briefly touched on by sashko in an earlier podcast and is also found in the overflowing github issue.
Planely: At design time of the package system it was simply forgotten to track for each package the underlying node version they run on. So with the update to node current, pretty much every package has to be re-build… mostly a logistical annoyance and improvement of the package system.
Super stoked@sashko helped line up 2 great guests for show 10. Meteor and Galaxy project managers, @zoltan and @rohit2b. The show will focus around free and paid Galaxy hosting, the business side of Galaxy and how free hosting getting axed has raised concerns for package maintainers and their demo sites.
@dinos I finally read the whole census post and PR. I totally agree with you. This was a missed opportunity and instead sounded some alarms. Luckily (and this is why I love ope source) there is time to take feedback, adjust course and still meet the requirements of both sides.
Probably the closest to this is @tmeasday. He’s been putting in a boatload of work over the last few weeks to make the docs, tutorials, and guide really stellar for Meteor 1.3. And there’s more coming after that - the docs (specifically docs.meteor.com) are probably the worst part of the Meteor experience right now, and we’re well aware of that.
I may be in a minority, but I actually like current docs page. Having everything on one page has its benefits, because its easy to find stuff. And links to method source are an amazing idea.
I think this is something we are going to have to compromise on, because the status quo right now has serious drawbacks:
It’s impossible to find anything on docs.meteor.com with Google (go ahead, try it!)
The page is incredibly slow on mobile because of the huge volume of content
I agree that Ctrl-F is really nice to have, but I think we’re going to have to figure out a different strategy for searching. Browser search works for libraries like Underscore where there isn’t that much stuff, but for a massive framework like Meteor I think the single-page approach just isn’t working out anymore.
Here’s the episode! Currently uploading to YouTube.
Sorry for dropping the ball on editing - we had a big change in the process, and I became unexpectedly busy. I’ll try to do better next time. Hope the content is interesting!
