How those quotes you made disagree with a statement that there should be more community involvement, but of course not that community should have complete control? I am repeating that MDG should try to empower the community, use the community, work with the community. Yes, I do repeat myself. Sadly.
And of course there are trade-offs to be made. And I think MDG is realizing this and is moving in this direction with some great steps recently. And I wanted to point out that here in this particular situation there might be a new opportunity for something like this.
And now I think enough about me, and let’s refocus conversation back to Meteor.
The market requirements you’ve raised are exact the reasons why Meteor/Galaxy can’t be a perfect solution for all markets right out of the gate. Remember, Galaxy has been in market <6 months! Don’t confuse the fact that we don’t immediately ship solutions for all markets as a proof point of our ignorance or lack of understanding of those markets; there’s a fairly good pattern of highly successful companies are those who are willing to say no before they say yes. So rest assured, we are well-aware of the growing popularity of Meteor pros in the life science/health care space and their specific requirements. We know this because we have direct product feedback loops with a substantial number of commercial customers (ranging from startups to large enterprises). But delivering VPC or on-premise support is not trivial, especially when we haven’t even delivered support for Galaxy outside of AWS US-East or a number of other features that solve horizontal needs for big swaths of professional Meteor devs like @lassombra who is a rare breed in this particular forum.
@lassombra Brian, you’re spot on when you conclude “that most “standard enterprise developers” are more in line with my thinking”. Can you PM me so we can chat further about your requirements and expectations?
Apple can afford to subsidize free app listings on their marketplace because it drives direct adoption (and loyalty) of a suite of entirely paid products/services. Meteor doesn’t have that luxury. If free Meteor hosting did lead to paid Galaxy adoption, we definitely would not retire the service or offer a direct equivalent. In fact, the proliferation of community packages is a headwind to developers new to Meteor. We’ve heard this loud and clear from the NPS analysis we do on all developers just getting started; they are confused by the myriad of choice and options of the JavaScript ecosystem. This single biggest request from professional developers new to Meteor want is more directed content/training/tutorials to help them build their specific app/project vs. access to lots of package demos to browse. (But this doesn’t mean we’re going to abandon great demos from top authors - we will work with you!)
So before you insult the MDG team and imply we’re an ignorant bunch of business people and that [quote=“maxhodges, post:41, topic:18949”]it’s probably beyond their resources to come up with any real solutions
[/quote], you should check your facts. Our business and product teams come with expertise from some of the most widely adopted and successful open source technology and SaaS companies in the world (Hortonworks, Mulesoft, Salesforce). We know a thing or two about building raving developer communities alongside long-term commercial growth. Our board, investors, and advisors also come from legendary companies like Spring, JBoss, Mongodb, and Docker. We make decisions with a lot of deliberate consideration based on our collective experiences and actual feedback from our entire community (not just the vocal ones that are here on this forum). In fact, almost all our top commercial customers don’t even participate on this forum because they’re too busy running their companies and shipping new features and products using Meteor!
So of course there’s a way for you [quote=“maxhodges, post:41, topic:18949”]
to “run” an app on the pay-as-you-go service, but force it to spin-down when there has been no activity for 5 minutes, and spin it up again on demand (like the free tier).
[/quote] But what clear business results does this achieve when compared to shipping things like multi-region Galaxy support or 1.3 or Apollo? With ~30 MDG employees, every “easy/no-brainer” project idea suggested on this forum costs time/money that comes at a direct tradeoff against every other item on our roadmap that delivers more value to customers and professional developers. Just because we decide not to do something doesn’t mean it was a bad idea or that we don’t care about the community; it’s just not the best option at the time when compared to the growing list of more important things we need to deliver to remain a viable player in the app development market.
So let’s get back to the heart of this thread. [quote=“maxhodges, post:41, topic:18949”]
What kinds of solutions can the community suggest?
[/quote] Well @mitar, instead of a generic policy survey that asks the community on forums how MDG should allocate its resources, the single most useful thing you can do is help compile a list of the most useful package demos currently on meteor.com. It’s up to you what criteria you decide is “useful” and as long as a number of your counterparts contribute/validate the list, I think we’ll be in a much better place, don’t you? If you can do that, we’ll be on our way to solving the original need raised (and that MDG wants to address!) on this thread. Are you in?
Thank you for your reply. But just to get this out of the way:
I do not depend on free hosting. I don’t need it. But my reputation depends on the recommendations I have made:
“Trust me, the Meteor Development Group is a highly accountable vendor.” <- Here, those accountability questions is what I’ll get. Along with questions about my professional judgment.
Trading trust. The top product teams of the world will not help, if communication and self-reflection is poor even on 180 decisions. Especially if it looks like MDG is suddenly short on money (bad), or short on a clean road map (worse). I really hope you guys learn to tip toe rapid decisions in feedback or option discussions with the community first. Meteorpad was run by a single part-time developer himself. I am really not sure whats going on resource, money, or story wise. But certainly more professionals read this than are vocal about it and it’s annoying that every two month stuff like this happens.
Regarding official partner:
I am in web tech advisory for B2B enterprises and start-ups. Here, being a professional partner of anything is not a really great way to signal objective stack assessments/cto assistance. Also, hosting in the US is just a no-go for 90% of my clients.
My opinion is that too many people saw the *.meteor.com hosting as a viable long term hosting solution. That was not the purpose or value for the free hosting. The community has won a lot from having the hosting but too many people abused it and put it in this position.
We need to change the definition of the free hosting and limit it severely.
So we, people who are on this forums, are a bunch of lazy bums who don’t do anything except for complaining.
Is that’s the “we know a thing or two about building raving developer communities” you are talking about?
For now, MDG doesn’t really show it that you “know a thing or two”. For now, MDG sadly works hard on a reputation of going more often against the community than with the community.
If you want to change it, by all means please do, I’ll keep my fingers crossed. But please don’t treat people like dumb and claim that you do things right when they see with their own eyes that you do not. Because I already notice people joking what will be the next big MDG announcement.
If you want to show that you do things the right way, don’t talk about it, but act. Prove people they are wrong about you by showing them that in practice.
I am what you could call a Meteor fanboy. I try to defend Meteor and MDG where I feel people are telling things against you and they are wrong. But you’re doing it unnecessarily difficult by showing that this bad reputation of yours doesn’t come from nowhere.
I’d guess most people here are sympathetic to tradeoffs … and that some inevitable bitter medicine would be easier to swallow if the tradeoffs were made more openly/explicitly, rather than behind closed doors.
MDG has done an incredible job improving transparency with increased forum participation, Transmission podcast, 1.3 early pre-releases, Apollo design doc, and so forth. A roadmap would really take things to the next level.
That’s not what he said. What he said was that they look at a different slice than is represented by this forum. There is intersection to be sure, but the forum does not represent well the group that MDG is targeting.[quote=“brajt, post:50, topic:18949”]
Is that’s the “we know a thing or two about building raving developer communities” you are talking about?
[/quote]
People keep confusing this forum with the Meteor Community. That is not the case. There are plenty of people on this forum who do nothing but stir up trouble and fiddle with Meteor on an occasional weekend 3-5 times a year. There are also more vocal professional devs, and there are tons of generally silent professional devs. This argument was what finally spurred me to be active on this forum, but I’ve been an active member of the meteor community far far longer, but in a much quieter professional space.
What the living hell? Seriously? Meteor has had a lot of useful announcements. They’ve done a lot of big things in the last year. They are getting more in line with the larger javascript ecosystem, which will end the complaints of about 40% of non-adopters (Why not NPM? is a very popular question on quora for example). They get one that is a 180 without explaining why it’s a 180. one poorly managed announcement and the entire forum is ready to hang the lot of them.
This has been a challenge for me also as a result of this announcement. My boss has been keeping tabs in general on meteor and he sat me down yesterday and said [quote=“my boss”]Are they losing track of what their brand is? Should we be concerned that this is a change in tone from MDG? Are we concerned that we might not get some of the updates we are watching for?[/quote] That last part is the big one. Might we not get some of the updates we are watching for? The reality is, no. All of the updates we are watching for are still on track. This 180 definitely smacks of poor communication, but they don’t tend to make decisions that are bad for the community.
Ultimately at the end of the day, this statement is the most telling. MDG has a focus, it just isn’t defined by these forums. There is certainly intersection, but they are looking elsewhere for their main product line. This is no different from literally every other SaaS provider out there. Look at github. They haven’t once added a feature to the free tier that wasn’t also a value add to their enterprise customers. And by the way, their enterprise customers are huge, and that is how they can afford to also provide the free tier. In their case they built the business free then paid but their paid tier is also much more costly than meteor’s paid tier so it did a better job early of funding continued free tier.
Bitbucket is an interesting anti-example. They have a free tier with all of the same features as their paid tier but with user count limits. That’s no big deal to really small companies, but the real purpose of that free tier is to sell Atlassian’s product management tools. They keep the free tier online as advertising, but have it so limited it doesn’t cost much.
Finally, lots of companies are now jumping on the “free for open source” bandwagon as open source makes their platform more valuable. In almost every case, their paying customers benefit from open source, so they benefit by making open source more viable. Meteor will probably jump on that bandwagon soon enough, but it will be done carefully and methodically. Gone are the days of meteor swinging their decision club around and taking home whatever they hit. They have evolved and are being more meticulous about their decisions. Some of them are unpopular, but you can be assured they are thinking things through, and doing so from the perspective that makes sense to them.[quote=“dinos, post:48, topic:18949”]
Regarding official partner:I am in web tech advisory for B2B enterprises and start-ups. Here, being a professional partner of anything is not a really great way to signal objective stack assessments/cto assistance. Also, hosting in the US is just a no-go for 90% of my clients.
[/quote]
This is prohibitive to a lot of developers who can’t have their name associated for one reason or another. Some companies don’t want the world knowing what technologies they are using. They see their stack as a trade secret. I don’t necessarily agree with that mentality, but I do work for one of those. Freelancers can’t be seen to have conflict of interest as that almost always will lose them the job.
We’ve been fairly transparent on the open source side of Meteor as you’ve mentioned; @debergalis’s latest Meteor Night talk is a great example of what’s coming directionally this year. However, like many software companies, we can’t be as publicly transparent with our commercial product roadmaps because it’s a competitive market out there! We do brief many of our customers on roadmap through the course of support conversations, small group product feedback mtgs, and also host larger customer advisory board events. Beyond our open source and commercial products, we also invest in a number of community-facing programs like Meetups, content creation like the Guide, and are seriously investigating putting on a first-ever Meteor conference. These may seem like small projects in isolation but are some of the many tradeoffs I mention that we load balance daily.
Sounds like untangling the relationship between Meteor open source development, which is all about working with contributors and integrating interesting technology, and the Meteor business, which is about keeping the lights on, should be a big thing to cover on the next TRANSMISSION @benstr. I don’t know if this kind of back-and-forth debate is the right way to assemble a coherent and correct story.
DAMN COMMUNITY DAMN, BACK AT IT AGAIN WITH THE COMPLAINING!
Look, Galaxy is a business. Theres no free lunch. Lets wait till things calm down a little before we get our pitchforks. Hosting is a freaking Perfectly Competitive Commodity business. If your company is just starting to figure out their product with some high-level customers, working out the kinks… WHY TAKE THE ADMINISTRATIVE BURDEN to worry about JOE SHMOS package that no one cares about…
I do care. People who use them do care. I know I am no one for you, just like I have no idea who you are. You may disrespect those who write libraries for others to use and those who use them, your life choices are yours to make.
But showing such a lack of respect will come back to you when you expect that the least.
I see that your post got a like from decisive people here, I take it that they agree with what you wrote, so I see no point continuing this discussion.
Perhaps the likes were related to things like: [quote=“abhiaiyer, post:56, topic:18949”]
DAMN COMMUNITY DAMN, BACK AT IT AGAIN WITH THE COMPLAINING!
[/quote]
and [quote=“abhiaiyer, post:56, topic:18949”]
Look, Galaxy is a business. Theres no free lunch. Lets wait till things calm down a little before we get our pitchforks.
[/quote]
Taking one little piece out of context and making the whole post about that is destructive to the community, especially when you completely missed the real point of the post with that bad quote.
Oh, @marktrang, what about this idea. Everyone can get a free one month of Galaxy, one instance. This is enough for you to setup a demo of a new package, and get people to use it. If after a month this is really something which is used and gets traction, then system detects this in the same way it detects this for current top packages and keep it free running.
Everyone already gets at least a full free month of Compact container capacity as part of the promo code MARCH25 when signing up for Galaxy. The issue is how do you recommend we measure “traction” and the system detect this? Atmosphere tracks package popularity but doesn’t have any insight into the package demo itself. And how do you solve for the rising star issue that @mizzao described? Again, I ask for your help - instead of recommending more engineering work to solve for the long tail of package demos, how about we start from a known list of must-have package demos that either most devs new to Meteor should see or that most professional apps use?
I’ll back it up, since from the very beginning I didn’t ask for anything more.
I don’t really see the need for the 1 container per package solution here. Managing 10 thousands of instances just for 100 demo/docs sites is too much of a hassle.
As much as they’re useful and needed, there’s not many of them and the number won’t significantly go up in the future.
Also the package doesn’t need a demo up and working until it is ready enough to be used in at least MVP projects.
“Guys, this is my package and this is Github link to the demo, let me meteor deploy it” in some topic on the forums is much, much easier to maintain.
Actually, it doesn’t even have to be Galaxy, the current *.meteor.com infrastructure is fine enough for docs and demo purpose as long as it doesn’t go down so often, and with only the demo sites, it won’t. After all for a long time it was pretty fine.
