A Sad Day (Layoffs in Engine Yard's Rubinius team)
blog.fallingsnow.net
blog.fallingsnow.net
Thanks! Nick Vella nick@rgatech.com
Passenger is great, but it's not an alternative to hosting with EY.
Was Rubinius a sort of R&D department at EY?
These layoffs are not as bad a news for a customer. It's just that as a hacker or an open source adopter, it is a bit of a blow. It is like someone taking back half the lottery money you initially won. (Hell, it was still a lottery only.)
There are very few companies (especially startups) who go so far as Engine Yard does/did in actually hiring people and supporting an Open Source Project to this extent. And they still haven't completely pulled the plug off it yet. I sincerely hope Rubinius does become a stable implementation of Ruby, and all the hard work pays off well for the contributors.
Disclaimer: I don't work at Engine Yard, and am thousands of miles away from the place of action.
I was the one that asked Ezra to keep an eye on Evan and Rubinius after RubyConf 2006. We really felt that the community would benefit from an industrial strength Ruby runtime, and still feel that way today.
We hired Evan more than 6 months before we received venture funding.
When we received our series A funding, we hired the best team we could get our hands on and hoped to push Rubinius through to 1.0, which we define as no apologies MRI parity, as quickly as possible.
Things are good here at Engine Yard. We have a growing customer base, and very interesting new products in the pipeline. As with all other business in these times, sharpening the budget just makes very good sense.
We did what we felt we must do to make certain we could continue funding Rubinius through to completion.
I've never once seen a company with a good hiring strategy for acquiring employees before actually needing them that worked. The fact that Digg has 60 employees is absurd. What in the hell does eBay do with 13,500 employees? A 4-person startup could rebuild all essential parts of eBay in six months.
Startups have shown over and over again that you can make great use of technology with very few people. As people are added to the team, especially in engineering departments, productivity starts to slow. There is a ton of overhead in integration, teaching and building company processes, and management.
Employees are toxic.
As an investor, one may wish for several related ideas to be explored concurrently with the hope that one sticks. As an investor, one may also desire that the business be prepared for aggressive sales when their idea (hopefully) finally goes big. Go hard or go home, so to speak.
I'm not arguing for or against these mindsets, just that if there was not common value in staffing large and fast, VC rounds would be structured like trust funds, not lottery payouts.
We're trying to build a tiny little mini-eBay in a tiny little niche market, and it's taking us fucking forever to do stuff. The issue is that it's really hard to determine what's essential - and to get a new seller (and we don't ask people to leave eBay; we just want them to add Dawdle as a venue), we have to be better than eBay on all those things that particular user feels is essential.
That's why Mixx, despite being a nicer community, with all sorts of bells and whistles, and links on lots of major sites, isn't taking off (http://siteanalytics.compete.com/mixx.com+digg.com/?metric=u...). There's a certain je ne sais quoi about a site that determines what makes it "essential" - and it usually isn't feature equivalence.
Building those essential things-- those things people absolutely need to use your service (and maybe to differentiate it)-- and then iterating upwards is the key. In six months, I think you could replicate the majority of eBay's functionality with four people.
The things you're discussing are mostly user acquisition problems. That problem is not solved by hiring more engineers.
Employees are toxic.
That is over-simplified to the point of being almost content-free. Beyond a certain point, adding more engineers to a team obviously decreases total productivity (per MMM); that's not to say that there's no value in, say, adding new sales engineers to support new customer engagements, or adding new employees to kick-start an entirely new project.
If you want to get eBay down to four people, you might have a startup auction site, but you won't have anything resembling eBay (for better or worse).