fyi they don't need to power cycle your machine to remove it from the data center (if there is physical access); there's a battery backup that clips onto the power prongs while it's still 3/4 in the socket.
update: I just tested this; if you're on 1.8.7, you can manually apply the patch I linked in the parent comment and recompile. There is nothing preventing a backport.
This article looks fairly accurate (for what I understand of the SF market) Things are a fair bit cheaper/easier in Portland. I do actually own a food truck in Portland (as well as a software company).
We made a profit on about three different days last year. The other 300 were not profitable :( It would have helped a lot if we had family members working for chips and hugs; paying staff is a huge cost particularly if you want quality gourmet-style food and hard-working employees (we have mostly 5-star yelp reviews)
Before our pod landlord decided to shut down abruptly, combined with our chef/my girlfriend getting diagnosed with brain tumors, it was costing about $3k/month (net loss) however that's the awful wet northwest winters driving people away from outdoor pods. Once the warmer weather starts back up again and we find a new pod, we should be in the black, but the business model I'm working on involves a lot of carts with a central kitchen, in clusters in a few cities. If anyone wants to chat about investing, let me know :)
We got started for about $20k total including the 'truck' which is actually a trailer that is stationary.
we just switched 1/3rd of our infrastructure off our existing host (engineyard, which uses AWS) onto raw AWS and saved about $2500/month. You can do it too!
It's not just a scheduling, there are a bunch of legacy systems you have to integrate with. And of course it's health data, so there are more requirements.
Anyone want to burn three months on an unpaid spec? :)
Elasticsearch is great and magical, but there are a bunch of defaults that you MUST set for it to be useful. I'm surprised github wasn't using these, actually (like allocating the min and max memory to be the same size).
Generally it takes a catastrophic failure under load for you to discover that 'everyone' (everyone else) uses these!
This is old news, but we successfully use https://github.com/freels/table_migrator in production not on heroku. It creates a copy of the table, performs the schema changes, copies the data over, then renames the tables for almost (i.e. 1-2 seconds) no downtime.
Knowing how long things took vs estimates is useful for determining the accuracy future estimates.
Having an absolute sense of numbers for types of projects is useful for estimating future projects of similar scope; where by absolute I mean, not, "I think it took two weeks last time we did this".
At entp we used a custom-built time tracker that works a bit like twitter; you tell it what you're working on (projects are referenced with @projectname and individual tickets with #1234) and it tracks the time between status updates to that project. It shows what everyone's latest status is, so it's good for a team to get visibility into everyone.
We used it extensively when we were mainly a widely distributed consulting company (4 countries, 4 US states, 20 devs) and continued using it after we dropped the clients. It's useful to see the hard costs of building certain features and even entire apps; there's also a ton of useful data about how many hours people are pulling a week, what time we show up for work, as well as the basic "what are you working on?"
However, the dev team is pretty small right now so we have ditched the time tracking. We are moving to a new tool that we're building which focuses more on how you feel about what you're working on.
"Launches"? Heh. I'm actually the lead dev on this (entp). It's been around since at least 2007, when my git history starts. Nothing new about it, but it seems to be a good way for Oracle employees and customers to communicate about products.
I've done this (both designing and manufacturing from scratch, and making t-shirts).
It looks like you mean "I want to design screen prints", rather than design clothes. This is actually pretty easy. Find a local screen-printer, and they'll probably be able to sell you some American Apparel (or Alternative Apparel) blank shirts at wholesale rates, and print your designs onto them. You might make about $10 per garment. It's really, really, really not worth the effort of having tshirts or tanks custom made.
If you want to truly design clothes; each step of this process requires hiring someone, since you probably don't know the skills or industry or have any contacts. Unfortunately, the newer you are at this game, the more people you have to hire, and the less money you can make. In fact, it's almost impossible to make any money at all as a new label unless you do most of the work yourself, and this involves experience and formal schooling, mostly.
You'll need to pay someone to sketch out your designs, and convert those into patterns. You might try doing this yourself, but this is a little like building a house without an architect. Or building a website without knowing HTML.
Next, you will have some samples made up, have the patterns modified so the fit is good (bigger labels actually have people called fit models). You'll need to adjust the pattern for all sizes you want to sell (XS-XXL).
It's best to have your samples made locally so you aren't waiting 6 weeks between revisions. Fit is everything! Once you have a few complete sample lines, you can take them around to local retail buyers and try to sell your line based on these.
You'll have to make sales before you actually have your garments made in bulk, unless you have a bunch of money to throw down the drain. Many people go to the Magic or Project trade shows in Vegas. You rent an (expensive) booth, make appointments with all the buyers, and on the days of the show, they come and make orders. (You don't know the buyers? You'll need to hire an agent) They won't pay you yet.
If you do make orders, now you can get your garments made in bulk in China, or wherever, and hopefully they can be shipped back by the time the orders are due (i.e. in time for the next season).
Something that takes half an hour to make will cost you several dollars in the US or tens of cents in China.
Unless you have a good contact at the other end, the first few runs will be horribly wrong, because people don't pay attention, particularly on new/small clients with low revenue. This happens no matter where you have things made.
Finally, you get your garments shipped, and then you send them off to be dyed, screen-printed, tagged, and packaged.
Then you deliver them to the places you sold to, and then 3 months later they pay you. Note that this is the first time you see any money in this whole process..!
It's worked twice already!