Meteor.com free hosting ends March 25, 2016

I’ve tried Meteorpad but it has never worked for me (at least in the UK). I get a constant “ERR_CONNECTION_TIMED_OUT”.

Why does nobody use it?

I am teaching a college course using Meteor for Mobile apps and our final project was going to use the free meteor hosting for our apps at the end of April. Is there any way this could be pushed back? Two weeks notice is very short.

I’m pretty sure meteorpad isn’t available anymore. I’ve read something about it on the forum and the github seems to confirm : Meteorpad

I’ll throw in my two cents, even though my thoughts largely mirror Peter’s & the general consensus

My situations that of the lead developer / CTO of a small company that is close to securing it’s second round of investment. Part of that investment has been secured by rebuilding the old site, ground up; and Meteor was my choice for a number of reasons (fast dev time, brilliant infrastructure for a small team, the looming promise of Galaxy) - it was the technology for tech startups right?

  • Our development has been erratic over the last 6 months, courtesy of a myriad of announced changes, which were often U-Turns of previous decisions. To my mind, 1.3 has brought too many breaking changes if you want to embrace the benefits it brings. I understand they may be necessary, but calling it a 1.x upgrade feels insincere.
  • The loss of the free tier is of no real consequence to us directly, but the way (and timeframe on which) it has been communicated is poor. As everyone points out, we can’t complain about a freebie being taken away, but the manner in which MDG operates shows a real lack of respect for serious developers (those that value the important things like reliability, clarity, consistency; rather than just “new hotness”)
  • This poor communication is a real concern. How am I to know if large, breaking changes aren’t going to be thrust upon us at short notice? It’s not an enviable environment to build a stable business in, and one that I am very wary of now.

I understand that MDG owe me nothing. I am using a brilliant (some of the time) technology for free, and that until now had a free testing environment too. What’s more there was a very passionate community maintaining brilliant packages; all backed by a company with lots of investment.

However, I now regret Meteor as my stack of choice. If it were an option, I would jump ship and start our rebuild over (and gladly give up lots of Meteor’s niceties like DDP); purely because I can’t trust MDG’s stewardship. They often make the right call…eventually; after a lot of indecision and heartache. Quite bummed out to be honest, as I had hoped that they would run their business with more diligence and respect.

11 Likes

We’re both circling around the same concern, but articulating it differently.

The “canary in the coal mine” analogy is apt, and I think one of many signs that the tech/VC world is in for a rough few years. Meteor isn’t the only place this is becoming apparent.

We’ve yet to launch, and were going to use Galaxy; purely because I’d rather we concentrated on product, not dev-ops, but again I’m now cautious and will be examining our options very closely.

5 Likes

I’m reading a lot of “MDG doesn’t owe us anything”. That’s bull. First, they wouldn’t be where they are without us. Not me personally, my contribution was a drop in the ocean, but us. “They’re a business.” Yes, they’re another business who builds themselves on the back of a community then bites the hand that fed it.

Second, they do owe us basic respect. Don’t take away with two weeks notice something you’ve been providing for years, especially with no migration path. Don’t break promises you have been making for years, especially overnight.

This is an extra bad time to step on the community’s toes, since all the effort at “aligning Meteor with the JS ecosystem” means it’s a pretty good time to migrate to something else. Which is what I’ll be doing now. It was already an uphill battle to convince people to use it for projects, and now the advantages are disappearing one by one, so why bother. Meteor up to 1.1 was a great breeding ground for the future of web development, and now it’s done its job.

To paraphrase the announcement: Thanks for creating Meteor and best of luck on your future projects.

5 Likes

@AndreasGalster sorry, but I disagree. And actually, I think there were some more occasions where it was stated that the free tire is not going away. And I don’t care about hosting apps.

I care about demoing meteor technology & packages!
bye touch2s.meteor.com

This is clearly a fast shoot into the foot and I would appreciate if MDG would keep their word on this “never goes away” @gschmidt. We all already do promote Meteor as a company. For free. Least to expect no one should go through the hassle to host & pay for demoing packages and meteor tools.

Meteorpad already died without any engagement to keep it alive. Which was hosted by one developer.

PS: I thought it got much better in engaging the community first after the blaze debacle, but apparently “announcing” the shutdown of 10.000 or so community showcases can still be done in a weak, thought out announcement without any resolution path.

4 Likes

Great point @juliancwirko. It would be great to read @gschmidt point of view on this. Does he really want all the open source examples to disappear?

I have been using Meteor to play around, try out ideas and for academic purposes. So I have deployed a dozen sites or so which I usually don’t use ever after. This means I am currently not in your commercial target group and will not argue from an commercial standpoint as some people try to do here.

Instead I’ll focus on a few other points:

  • you are breaking the ease of deployment for trying out things. For professional developers that are deploying apps this is certainly a non-issue. For getting an idea, firing up a meteor app, scribble a few lines of code, deploy and check the result on your mobile, it is.
  • you are loosing a big wow factor in demonstrating Meteor for workshops, hackathons and the like. Meteor used to radiate “magic” compared other frameworks. Being able to deploy from “within Meteor” without setting up another account or service in this way was part of this magic.
  • you are loosing people’s trust by breaking the verbatim written promise of a free tier that will never go away
  • you will loose people at the free tier. If somebody is supposed to set up / configure two different services just to get an app running with Galaxy, and people tend to go for free (like heroku and mlab at the moment), why should they switch to Galaxy later?
  • you will now loose new people at the very beginning. There are lots of tutorials, videos, blog posts and books out there and this material changes slowly. After March 25, for those people the first contact will be a disapointment when they find out that meteor deploy doesn’t work the way they think it does.

It should be technically possible to enable a free try out sandbox without cutting into your business and I would appreciate if you would be trying to figure that out.

11 Likes

That’s a pretty important thing.

5 Likes

Does it impact users who use meteor only in mobile application? Will I’am able to build and deploy mobile application?
Thank you

Should Galaxy and MDG PR be Transmission #10?

12 Likes

I wrote an article on Medium to share my thoughts about this. To me, at least, a bit of the “magic” of Meteor will now be gone…

8 Likes

Yes. Get on this ASAP.

It’s really not that bad of a situation, but the rumor cycle and uncertainty is going to continue until the service shutdown. Put *.meteorapp.com high on the list of things to discuss as well.

7 Likes

It comes down to this… tech companies offering free tiers are disappearing for good reason. MDG has been an exception to that rule, offering both free hosting and free premium OSS. I never imagined both being able to stick around forever. I am glad to see free hosting dropped since their investment in OSS is their real value offering to the community.

Will the magic of Meteor disappear because free hosting is gone… no way.

Did MDG miss a great Galaxy PR opportunity and instead spurred distain… unfortunately it looks that way :frowning:

But we do not know the whole story, and assuming things is always a dangerous bet… I am hoping to uncover the behind-the-scenes story on Transmission.

20 Likes

I have a feeling we will ultimately find out more about this decision and it will seem more understandable.

I do have a serious suggestion for MDG/@gschmidt; you could consider letting a few or the long-standing community members see these big-ticket press releases before posting them. The community’s concerns are different to MDG’s and if you hear those before you post it would allow the original press releases to be tweaked to answer all the big questions and uncertainties before they ever arise.

That would hopefully stop these big storms of bad PR that I imagine don’t help Meteor’s reputation in the wider JS community.

To be clear, I’m not saying that MDG should change the actual intention of whatever the particular message is, I run a business, I get that you need to do what’s best for yours, but I do think there’s room to improve how these messages are communicated initially by hearing all the views involved at the outset.

10 Likes

Just to throw my two cents into this hot topic…

When I made the decision to take web dev more seriously, I mentally set aside a certain amount of money every month as a budget. I think that in comparison to other hobbies/trades/etc, the web dev community has things pretty good – free tuts by the thousands, open source ides that are maintained by the community…

You can really get started with web dev for $0, assuming you have a computer.

$50/month for me to maintain a handful of droplets, pick up a domain name now and again, pay for a bucket on amazon seems pretty reasonable for someone making the transition from hobbyist to something more serious. And I think $50 is high.

A $20 droplet on DO could definitely host quite a few demo apps, and you can grab a domain name for under a buck. I guess if you’re a weekend warrior who programs just two days a week, you’d be looking at around $2.50 per “session” – when you think about materials, I think that’s somewhere in the price range of cross-stitch.

All of that being said, I do think the way that this was handled was flawed, at the least. I think that with the way that mdg has been opening new channels for communication, there was definitely an opportunity to communicate this message/lead up to it in a better way. This felt like the text from a boy/girlfriend that basically says: “Sorry, things just aren’t working out”.

Although this decision/move doesn’t impact me directly, as I transitioned almost immediately to my own deployment options, it does give me cause to worry about things like blaze. With transparency constantly on everyone’s minds around here, this seems like a poorly executed move…

3 Likes

Yes, of course, I expect to have to pay for professional grade hosting, but not just to try something out.

Guess you’ve been fiddling around with your thumbs until Meteor came along because meteor was magic and nobody else was?

Where to go from here
As for me, I will continue to use Meteor for the time being, but now that I will be paying for even the posting of a quick test or demo app, I will be looking around much more actively for other options.

So setting up an entire mean stack yourself, then painstakingly deploy to the free heroku is fine but filling out the mupx file for deployment on heroku which literally takes 3(!) minutes after you’ve done it successfully once is … terrible? And now all of a sudden requires you to switch to the former option?

Sigh I really wish people wouldn’t create all these mimimi posts that then outsiders read and start to think really bad about meteor due to the lack of perseverance of the people making these posts.

MDG, we still love you, even if you really suck at communicating your stuff. But for god’s sake please discuss these things within your team and within a small circle of the community first. It was to be expected that shutting it down within 2 weeks (even if it doesn’t matter since that’s just the hosting, not the db) would cause a shitstorm.

11 Likes

At least make the galaxy free for quick deploy and testing of apps for users like 100 or so… its kind of major turn down for me … my true feeling right now is disappointed and sad

2 Likes

This would be easily resolved if meteor would come with boilerplate mup.json files for the major free hosting providers. I have always found it mind boggling that there are no easily findable presets for aws, azure, google cloud, heroku, etc

4 Likes