The first 90 percent of the code accounts for the first 90 percent of the development time. The remaining 10 percent of the code accounts for the other 90 percent of the development time.— The Ninety Ninety Rule, Tom Cargill, Bell Labs
The campaign was being run by author, pundit, and former InfoWorld columnist Robert X. Cringely and his three pre-teen sons and the plan was to take some off-the-shelf hardware, some available Minecraft server software, combine it with an admin interface of their own, and create an easy to use Minecraft server that you could setup at home and that mom could admin from her iPhone. The Cringely team had been working on this for a year and it was all done save for some details.
The campaign ended on October 21, 2015 with 388 backers pledging $35,452, well over the $15,000 the campaign sought. Thus funded, the team was working towards wrapping things up and shipping out units to customers by the end of the year.
Full production will begin at the start of November and our goal is to deliver all Mineservers™ — burned-in and tested — by Christmas.
-Mineserver Campaign Page
The final line on the campaign spelled out the attitude towards their goal.
This is not rocket science… That’s next year.
-Mineserver Campaign Page
So it was off to production. There were some updates through October and November about finishing up the admin software and sourcing parts and getting units ready.
And then things got a bit quiet in December, until on December 21st there was an update with the title “Man Overboard!”
Their Linux consultant, needed to tie the whole package together, had gone missing and had to be replaced. He was a rare person, being a very familiar with Minecraft, but they found another guy who was very familiar with the hardware, but wasn’t up on Minecraft. Shipments, already a behind schedule, would be delayed a bit longer.
Again, we’re sorry, but shipping will be delayed about one week.
-Man Overboard update
And then began the long dark winter, as more than a month went by before the next update titled “Better Late Than Never!” More problems had cropped up. The compiler they were using for their chosen version of Linux wouldn’t work with the ARM based motherboards they had chosen, motherboards which they had purchased and were waiting for installation.
This was going took a bit of time to solve, but wasn’t insurmountable. The custom admin software was moving along and they had support for the server flavors they wanted, except for Cuberite, which they expect to add soon.
While they worked on Cuberite support and a WiFi issue, they put up some test servers so people could try them out for performance and use the routing software to see if it resolved correctly. (It never did for me, I had to use the IP addresses to connect.)
Comments from backers, which started to take a negative turn after the last gap, got a bit darker when a month and a half went by before we heard another word from the Mineserver team. On March 17th we got an update titled “The Devil is in The Details.” There was a statement up front that they won’t delay between updates that long again.
Getting the Cuberite server software to work with the admin application had been problematic, it being the odd man out on the Minecraft server front as it is compiles in C++ rather than Java. This however was what made Cuberite desirable, as it is way out in front of the other options in terms of performance and really needs to be on the box for the Mineserver to meet its supported players claim.
Meanwhile, the dynamic DNS system and a the WiFi support also had issues which needed to be solved before they could ship. And, finally, there is what they think is an Ethernet issue causing the boxes to drop off the network occasionally. But the mood was still upbeat. These were solvable problems, solved already really save for the last one, so the update ended with optimism.
It’s the final bug, we’re approaching it with planning, gusto, and plenty of Captain Crunch, and fully expect to solve this last issue and start shipping next week when the kids are off school for Spring Break.
-The Devil is in The Details update
Three weeks later there was a small update that included a call to help test the Mineservers they have setup and a mention that the Ethernet bug may not yet be solved.
Then there was another month and a half gap, proving that they in fact would do “that” again, before we got the Science Experiments/Finally Nearing the End update. Cringely’s son Fallon used the Mineserver setup for a science fair entry, which features at the top half of the update, which then goes on to the fact that the underlying management software still doesn’t support the Cuberite server. Support for Cuberite will be available at some point in June, so they are going to wait for that.
Meanwhile Mojang has released Minecraft 1.10, which comes up in the comments as to how updates like this are going to be handled. There is no response to that question or, frankly, any question ever that comes up in the comments. Cringely doesn’t do comments, he just drops updates and goes away it seems.
An early July update said that the Cuberite support had been delayed, but that they continued to work on the product and that it will support Minecraft Pocket Edition at launch, which needs its own flavor of server. There is a promise of regular updates.
At this point we’re going to start doing updates every Thursday with the idea of keeping you better informed. Thank you for your patience.
-Still waiting for AIM update
By July 22nd there seemed to be a breakthrough, with an update that says Cuberite support is now a thing, and getting it to work right is all that is delaying shipping. Optimism still reigns. They have only missed one Thursday update since making that promise Again we are told, “This isn’t rocket science. ”
The next Thursday update has an admin interface issue solved, but there is still the Cuberite support issue. A week later, on August 5th, things seem really close, plus now the Mumble voice server software will be bundled in with the Mineserver.
And then summer vacation ends and Thursday updates cease. I expect that with three young boys heading back to school that the Cringely household has its hands full.
Finally, on September 14th we got an update that says the WiFi doesn’t work reliably. WiFi hasn’t been mentioned since January, and the comments indicate that some people have been taken aback by this. (One person claims to have filed an FTC complaint.) Apparently the part they chose isn’t compatible with Arch Linux, so they have ordered a replacement that should fix the problem.
No further updates have been posted, so the status of the WiFi and Cuberite support remains a mystery.
On September 23rd the one year anniversary of the launch of the Kickstarter campaign passed, and on October 1st it had been one year since campaign met its funding goal.
Welcome to crowdfunding.
Ah, but a project’s reach should exceed its grasp, or what’s a Kickstarter for?
-Me, abusing Robert Browning’s quote
Last week over at Massively OP there was a post about 10 questions you should ask before backing an MMO Kickstarter. In the comments I, along with several other, proposed some additional question, mine being focused on your readiness to accept that any dates promised during the campaign are generally blind optimism at best. I mean, Star Citizen, right?
Being late is part of the experience.
About a year and a half ago I reviewed the Kickstarter campaigns I had backed, and those that had shipped were universally late. Not all drastically so. Defense Grid 2 shipped just a month after the promised date. Go, go Hidden Path Entertainment!
In fact, I have been meaning to do a follow up on that post as time has passed and a number of promised delivery dates have gone by, but I was waiting for the Mineserver issues to get resolved, since it seemed like that would just been a few weeks off… for the last nine months.
I am not particularly annoyed by the delay. This is not a rage post, but a summing up of the tale so far. If I want something delivered on time, I order it from Amazon. I did rather optimistically plan to move our Minecraft world to the Mineserver at a couple of points, but not being able to do that hasn’t really changed much. Instead it got moved to different hosting providers.
This is more of a review post to look at how things were handled. These days I am more interested in HOW people run campaigns as the campaigns themselves, having come up with my own success predictors and such.
The campaign itself went very well. It got enough publicity, in large part because of the Cringely name, and hit funding milestones that indicated it was going to be a success early on.
No, the problems here have all been post-campaign. Success is a problem everybody wants to have, but how you handle it can make any victory Pyrrhic. For example, No Man’s Sky was a huge financial success (yes, I know, not a Kickstarter campaign, but work with me here), however the gap between what was promised and what was delivered will likely haunt the studio and key devs going forward.
For the Mineserver campaign I think there have been two obvious problems.
The first has been over-optimism, which is ever a curse here in Silicon Valley. At several points during the campaign it sounded like problems were just about solved and we were told that units would be shipping soon. And then they didn’t ship.
You cannot foresee all problems, and people will forgive you a couple of slips. But after a while your predictions lose their credibility.
The second problem has been communication. Updates have been sporadic and the comments might as well be turned off for updates section since they never get any sort of response.
At one point Cringely said he didn’t want to do updates unless he had good news to share. In my experience though, any update is better than no update. And that is doubly so when combined with optimism, when an update says there will be a week’s delay and then the next update doesn’t show up for a month.
I think the campaign was on the right track for a bit with the “every Thursday” updates. But, as I noted, those fell by the wayside and now we are in limbo again, waiting for some news.
So we shall see.
At this point, with our current server happily chugging away on Mojang’s hosting, I will probably use the Mineserver to play with Minecraft mods. When it finally arrives.
Addendum: We got an update on October 7, there are new problems, a possible solution, and shipping is still a few weeks beyond the horizon.