Why not? What would we get?
A setup of stateless servers/containers with little monitoring (what I see from the screenshot and what I read from the posts here), weak log support and a general 1h sla support that does not include my application itself and no support for the mongdb (because I have to host it somewhere else).
If something goes wrong we need a SysAdmin/DevOps on our side to make the error message good enough, so you guys can figure out what is wrong… Bonus: you can always blame the mongodb, So what can you actually support here?! Reboot a container? Spinning up a meteor server in a (docker) container and adding Kadira is done in a few minutes. A “reboot on non responsiveness” seems not so difficult to setup.
Starting with 650$/month (he, its cloud time - I don’t need yearly offers anymore) for 10G RAM and an incomprehensible number of CPUs with a service that can only work very generic in the gap between the mongogb and the app logic - I have really no idea why that should be so expensive.
Coming with smaller offers later, because you want to concentrate only on a few big players - does this mean you have not even setup an automated deployment and monitoring?! Having your cloud only “on the other side of the world” (Europe here) is not a good sign either… This all sounds like you started too early with too less and you urgently want or need money…
This all is scaring me.
Meteor is a great idea, the eco system and everything is good, so far your advertisement and community building was great - but now you reveal plans that give me the feeling I’m accidentally stepped on the dark side… Please don’t prove me right and fade out “meteor build”. How long will it take until this command does not get the newest features, so you get through with these prices?I have the slight feeling that the “vendor [wants to ] lock [me] in”…
(Don’t get me wrong. We don’t mind paying services and some services we pay for give us a great value for money, but I don’t see that here…)
Wait… what? They didn’t give you some sort of preview yet? Allrighty… that’s just plain silly (read: bizarre!) in my book.
I’ll refer to the other topic about the disconnect between community and MDG then
P.S., not ironic, it feels weird that someone with this kind of presence in the community is not in some sort of “loop”. If it’s because Kadira’s functionality overlaps with Galaxy’s in some regards, that still makes it weird
two separate thoughts woven into One Single Thought:
Will MDG be able to give the investors the $30 million they raised like this?
I am afraid & concerned that such a great, revolutionary Framework/Platform will die because it will run out of funding as it couldn’t generate enough revenue for it’s investors.
I earnestly, from the core of my heart hope I am wrong.
Thanks for the inside scoop @maxhodges. Part of being ‘a great devops’ resource for a shop, is setting things like log archival, log indexing and searching (For example, and ES/Kibana solution), and dashboards. It’s far more than the mechanics of scaling, obviously.
And of course, you need more than momentary data - seasonal Enterprise customers for example compare to same month last year, etc… So your points 2) and 3) really seem to point to a lack of understanding of what they’d have to do to truly replace Just In Time devops, as most people are doing.
one-click scaling and orchestration of deploying new versions across multiple containers isn’t exactly functionality you get out-of-the-box when you provision a new server is it? If it were that easy, why not compete with Galaxy and setup your own Meteor hosting company?
I think they understand. It’s an minimum viable product so they can start getting revenue to make their investors happy. They are adding features as they go, but at this point they thought it would be valuable to some companies to start selling it. (There were no stats in there at all until a few days ago!)
As a startup with zero income I could NEVER affort 500 bucks a month. NEVER!
I was hoping to find a plan for 30-50$ per month for good meteor hosting, meaning that I can run complex requests against mongo on bigger data-sets and still have a good performace and seo.
Well, we are lucky that there is modulist giving us another option.
I’m from the Startup Nation, which is 2nd to Silicon Valley and I can say without a doubt that 90% of the startups here would never pay that much for hosting at least when they’re in the first 2 years.
To all the startups who are objecting to the costs, here’s a question: is it a good idea to be horizontally scaling a website that has no revenue? It sounds like you simply want other people to foot your electricity bill.
Which is fine and all. But there are specific people who handle debt-financed growth of businesses: they’re called venture capitalists. If they’re not on board yet, maybe concentrating on developing a business plan that generates money is a better use of time than horizontally scaling an app?
MDG still providers a free tier. This pricing simply asks that people have a business plan in place and are generating some revenue before using the horizontal scaling infrastructure.
Also, news flash: startups aren’t the only market that MDG is targeting. Folks in the startup community have been feeling a lot of love for a long time; and those of us in the B2B market have been feeling left in the lurch. This pricing point is a pivot, and showing that they’re moving away from only concentrating on the startup/unicorn market, and looking to establish a client base of small/midsized companies.
If they integrated Kadira, MongoDB instances, CloudWatch-like alerts (and probably a lot more stuff which I can’t think of), I’d move over to Galaxy (for a price range of 20$-100$, considering that my needs in terms of power are low). That would be a service that would be miles ahead of anything else and I’d be willing to use it and lose some of the flexibility that spinning up my own instances gives.
Honestly, I’d just love to see MDG talk with @arunoda about integrating Kadira into Galaxy. @arunoda has the knowledge and experience in this area and a pretty good product already. Kadira on its own would still serve a purpose for people/companies that want to spin up their own instances, but if Galaxy were the source-of-truth for everything about my app, then that’s when I’d start really considering Galaxy. Right now, it feels like it limits my options and costs more. But I see huge potential for Galaxy, it just needs to go that extra mile and deliver full functionality.
I’m under NDA with most all of my clients, so I can’t list specifics. But I can give a market profile of typical businesses who contact me about professional services:
~ 5 to 50 employees
~ $1 to $10 million in annual revenue
~ 100 to 1000 estimated clinical/business users
~ 10k to 100k user/client records for the project they’re working on
most have a brick-and-mortar office
most have an established IT systems already in place
most have some type of professional licensure underwriting development
all of them have some type of pre-established revenue stream
most are building apps to help them expand into a new market or improve internal efficiencies
Pretty typical small-business demographics. Folks who are established in one area already, which gives them a bit of a safety net and some resources to work with; and who are seeking to grow. Typical clients that meet the above demographic include:
independent physician practices
nursing homes
laboratories and pharmacies
university research groups
hardware manufacturers
established service providers
For those of us in the B2B market, this is pretty typical fare. As far as applications that these clients are interested in, I’m seeing interest in the following kinds of systems:
practice management software
medical home software
hardware controllers and operating systems
scheduling and logistics (ERM)
e-commerce markets
clinical-trials coordination
And that’s just in the healthcare market.
Think of it in terms of multi-tenant based pricing. Instead of one megalithic Facebook/Google style app; this pricing is really geared to having a Wordpress-style application template that gets customized on a per-business basis.
I think the bottom line is that most meteor developers are relatively small-scale. It appeals to the ex-PHP crowd. So most people don’t care about all these metrics and scaleability; they just want to be able to type “meteor deploy” without messing around with a bunch of set-up factors and the thing works. That ease is probably worth $10-$20 a month for a lot of people, but not 25 times that.
Frankly, I don’t feel all that entitled to having meteor provide cheap hosting AND a free framework from which I can build the entire backbone of my online projects. If anything, it makes me appreciate how cheap modulus is, so it’s good overall for the industry.
