not sure if you have read my issue post on meteorpad queue, but I absolutely respect @mike s personal choice. The only part I am asking for and I guess that should be naturally and would be more than fair enough :: just to get a mongodump or whatever from the public sources.
So as @waldgeist already mentioned, there are a lot of answers (even some of my owns) on stack exchange and here in forum, reflecting to sample codes on meteorpad.
No one will direct Mike to continue work even not for free, but as long as he had promised the service I feel it is a good style to give people back what they have brought in.
Meteorpad was pretty rad, especially when I was on non-traditional devices ( an iPad, a Raspberry Pi… ) So I want to join in on saying thanks for all the years that Meteorpad operated. It was very handy and clever.
Now, was it “not nice”? Well, for a lot of web sites that get a little bit internet famous, there is a pattern. You might even say that there’s site-take-down etiquette. It would look something like this list, from the most common steps to the least common steps:
Put up a page that says that the site was taken down. The most obvious step to take because up until I found this thread what I did for a few weeks was try it, wait ~5 minutes for the server to time out, and then try again in a few hours/days. I did not know what was going on. It’s not obvious that the site is permanently down. Sites go down temporarily all the time.
Shameless self-promotion. The most common prose on a take-down page is fluff that reminds everyone who made that site, how much work it was, etc, with a links to everybody’s linkedin. Or even just something for anyone wondering… “What’s @mike up to instead then?”
Offer an alternative solution. which I think would be to just bite the bullet and install Meteor normally on a “real” computer.
Offer a way to get in touch to download one’s personal projects. Not everyone had ten dozen todo example apps. I had a handfull of really great projects I tinkered on from time to time on Meteorpad, foolishly not stored anywhere else. RIP.
Point to a roundtable Occasionally, the ex-admins will give a link to a facebook page, or a forum (like this one), or a hashtag where people can reminisce about all the good times and what fun they had. Boy howdy, what fun #Meteorpad used to be.
Can an object’s creator destroy their creation at their whimsy? Absolutely! And if it is well-liked, should they do so without leaving a trace? I don’t personally recommend it.
But it is not common in cases like this for creators to just give their creations away. More commonly, someone else will try to copy it.
Bad drop down experiences of free projects like that (especially missing 1. and 4.) will keep users away from using next upcoming ideas from innovators who are willing to handle everything in the common way (1. - 5.)
And for that I can’t understand that MDG is not interested to support the developers at least to get back the public project sources they have entered and also save that overall meteor knowledge. But maybe they do and dont’t get response from @mike as well? I don’t know.
Good to see some of the respected members of our community are just as concerned about the disappearance of MeteorPad as I am.
The REAL REASON I started this discussion was to put the spotlight on MeteorPad or even MadEye as, possibly, the next best project MDG can invest in. After reading the RoadMap, I do feel that this can come after Apollo!
First of all, I’m sorry for being silent on this thread and the similar GitHub issues. I’ve had a lot going on in my life the past few months, but I should have made the time to at least say something. Hope this is better late than never.
First let me give some background on how MeteorPad came into being. A couple years ago now, MDG contracted me to build MeteorPad. After it was built, they continued to pay for the significant hosting cost until early this year. Once that happened I wasn’t able to personally afford the hosting costs so I pulled the cord on the service. I want to emphasize here that I’m not upset with MDG for their decision. They’re a business and the hosting costs for the service were non-trivial. I’m incredibly grateful I had the opportunity to get paid to work on a project as cool as MeteorPad.
Like a lot of you I’m sad to see MeteorPad go away. I’ll have some free time (but not a lot) over the next few months and would happily work with people to keep some version of MeteorPad alive.
Option A is to modify MeteorPad to simply not run a server process per pad. This would dramatically reduce hosting costs. Unfortunately it also makes MeteorPad little more than gist for meteor. It would “unbreak” all the broken links to MeteorPad on the forums and stackOverflow.
Option B would be to open source MeteorPad (I currently own the source) and allow users to run their own MeteorPad servers.
Option C would be to try to raise enough monthly recurring revenue to cover hosting. I’d still want to open source MeteorPad under this scenario so I could have some help from the community with maintenance.
I’d be really interested to hear what everyone here thinks. In particular:
Are you willing and able to contribute to MeteorPad (technologies include Meteor, Docker, Ansible)
If I followed through with option B would you be interested in hosting your own MeteorPad server?
For Option C, would you will be willing to personally contribute money to keep MeteorPad running? I’d definitely be open to corporate sponsorship in this scenario.
I am always happy to contribute to stuff like this!
Just so the community knows what it is asking for, rough order of magnitude how much are we talking per month? $500 … $5,000 per month?
I know MDG said the free Meteor hosting was around $10K per week… I suspect MeteorPad wouldn’t be that extreme.
Do you think Digital Ocean might sponsor part of it?
I saw a thread where they suggested to reach out to them via sponsorships at digitalocean dot com.
That’s definitely an idea. It was hosted previously on DigitalOcean so maybe there’s some good will there. I’ll go ahead and reach out to them. Wish me luck
@mike so pleased to hear of your plans and boy am I excited of the possibilities! I like all the options you put out there but I’m leaning more towards Option C.
I think you have a great product there and I do believe you can raise that monthly recurring revenue through different paid plans. I, personally, would be willing to pay something in the area of $10/month for some 10 pads. (This being the starting option).
Above all, I’m 100% behind in making MeteorPad open source. I’d definitely be a contributor
The funny thing about is that nobody from MDG has responded to this thread while knowing that they’ve canceled the payments for MeteorPad. Just thought that we had a discussion about “clear communication” a while ago…
thanks for your feedback and open words. In fact that a lot of people want to see MeteorPad alive, I am sure you are able to get the contribution you need.
I will bring up my 2 cents to your posted options:
In general this is a good idea also for some ongoing ideas but I think there should be also at least this central running instance which is used by all as (today) to link from StackExchange etc.
I am not sure how much ressources, system load and traffic meteorpad has token. Do you have some benchmarks oder measurements about that?
However, have you discussed with MDG to get a free plan for running that on Galaxy?
On the other hand, if someone had to pay for Galaxy, what pricing plan do you think would be necessary?
We have built up a lot of experience with running our meteor apps on Scaleway ARM servers. Speed was always good and pricing is top. Maybe we can have a short discussion about running on such an environment.
For my personal opinion - Option C should be the short track - to get at least for the next 6-12 month a running MeteorPad environment. During that time it would able to make the right future decisions for you and all potential contributors.
Option A is also ok to me, even if it just runs as gist but unbreak the currently dead links. In case that this will take some development time, I assume still option C as the fastes reanimation.
In any case thanks for being back and opening the discussion.
For a $1000 a month, a lot of people would be willing to help pay that, myself included. If only 100 users were willing to pay, that would be $10 per user. And there’s definitely more than 100 that would be willing to help pay for this.
Some people will be more than happy to just donate their funds. But to play devil’s advocate, I know there will be some people who want something in return. So what should that return be?
If it’s directly user-funded, you’d need some governance over things like issue resolution, feature requests, future scaling measures, cost transparency etc.
Alternatively, it could be a simple donation-based scheme (via patreon for example), or a traditional subscription offering with a paid level of service (perhaps based on higher performance or persistence options etc).
While I am optimistic that the community could foot the bill for the hosting, I think it may make sense to actually see what people are willing to contribute. Maybe dollars and time per month?
So that we (@mike) can actually have some numbers to look at?
Some great ideas here. I’ll happily chip in if it goes live. I’d love to see Meteorpad and also hassle free free meteor hosting come back in some shape. How about a time to live on each app and people can chip in if they like it and make it run a little longer?
First off thanks everyone for all the positive feedback! Some quick thoughts on some of the points you guys have raised so far
@cstrat I just emailed DigitalOcean to see if they have any interest in sponsoring. Hopefully, I hear back from them soon.
@tomfreudenberg I agree that Option C (raise money continue to host) would be the easiest and the best. But it’s definitely not the cheapest. I talked with MDG about using Galaxy a while ago; it’s not really a good fit since MeteorPad has to run apps in development mode. Thanks for pointing out Scaleway servers; it looks like that could dramatically lower costs.
@merlinpatt I’m a little more cautiously optimistic that it’ll be easy to get to $1000/mo. Do you count yourself amongst those who’d be happy to pay?
@batjko Something like Patreon seems like it could work pretty well. I’m not really confident it would work as a paid product, but if you guys have ideas for a premium feature that would be compelling for you I’m all ears.
@vigorwebsolutions Thanks for laying out what you’re willing to contribute. Now I just need 99 more of you I think I’ll put together a short Google survey just to get a rough feel for how much people are willing to contribute.
