johnjones, we took inspiration from Quibb whose founder told us that the feature has worked well for her. There's something inherently cool about someone valuing your opinion so much that they ask you to come in and join their conversation, especially if that conversation is about something you're very passionate about, like your favorite sports team.
I was in Online Services (aka Windows Live) about two years ago. Some groups are obviously better than others, that can be said for any company. I think the problem here is that the author does not realize that unlike college, corporate world requires you to be a bit more proactive. Large companies may suffer from bureaucracy but on the upside, they have vast resources and smart/accomplished people you can tap into if you check your ego at the door, don't expect everything handed to you in a silver platter and tone down the poisonous attitude.
I spent 5 years at Microsoft as a dev and share almost none of the author's sentiments. I had very few meetings, my manager went out of his way to make sure that I was not blocked, code reviews were mandatory, blogging on technical aspects was ok, had ample free time to work on any side project I wanted (even encouraged to work with MSR on things that were more researchy), etc etc... I could go on for a while.
Sure, there was a huge emphasis on shipping rather than spending days space architecting the codebase and sure, not every piece of code was a stellar example that would show up in a college textbook but I'm having the exact same problems while doing my own startup, where at times I knowingly incur some technical debt in order to ship on time.
I was born and raised in Cyprus and my entire family still lives there. I spoke to a few of them this morning and whether the measures pass or not (parliament still needs to approve) they all plan on making a run to the bank asap.
I practiced writing code a lot during UROPs (research opportunities available to MIT undergrads). While UROPs are not required, most course VI (cs majors) that I know did at least one during their 4 years there.
I dealt with this exact problem for a number of years. Background scrubbing takes away I/O resources and can be a disaster on your workload if you rely on sequential reads/writes. For that reason, most controllers are configured by default to only scrub when the disk is totally idle which is never. Even if the controller had a better definition of idle, scrubbing an entire disk to find those rotten bits would take a long long time, a disk would almost certainly fail before that.
Distance between Cyprus and Syria is < 200km. How much money would it take to setup a wireless link between the two? Wikipedia tells me it's possible (http://en.wikipedia.org/wiki/Long_Range_WiFi#Italy)
Apologies if this is a dump question, but isn't it trivial to derive the public key if you have the private key? If so, couldn't an adversary just keep a list of public_key->service mappings and profit by casting a wider net?
It may be because I'm cranky this evening but I found the author's sarcastic quips really fun to read.
On the content itself, it sounds like if you're not targeting a specific service, you could code this up and launch a few hundred c1.xlarge spot instances (larger instances => fewer neighbors) and just collect keys. Or am I totally underestimating the amount of manual work required post-processing?
I know there are lots of smart people working there but just look at the sheer amount of AWS offerings. Amazon certainly gets credit for quickly putting out new features and services but it makes me wonder if their pace has resulted in way too many moving parts with an intractable number of dependencies.
I'm surprised that Redis cannot handle your traffic. What are you throwing at it? Or are you on EC2, with frequent BGSAVEs or AOF enabled on an EBS volume?
We are not using anything fancy, it's just Google Analytics, redis and postgres. If we send anything out to the world (email, link to partners, open graph action) we tag it either with a google analytics campaign or a unique identifier so we can trace it back to our database. Every 12 hours, using a bunch of SQL commands + API calls we get a report in the mail.
That is a great question. We spent a lot of time designing our signup flow and are currently on the third major iteration. The one thing that we find works consistently well is to give the user something to do as soon as they sign up. For Sidelines, that something is selecting your teams. The drop rate on that screen is extremely low and we speculate it will approach zero as soon as we add more sports. We are also working on implementing a "tour" of the site as part of onboarding, showcasing the various features because we did get the feedback that things can be confusing initially.
Hi, I'm the CTO of Sidelines (startup mentioned in the blog post).
For our landing page tests we use Google Analytics which now includes Google Website Optimizer. We track engagement by logging activity time stamps per user in Redis. We also track how many other friends a user invites as a means to measure our k-factor and we are definitely biased towards targeting users that are more likely to be engaged and bring other users to the site.
You're not alone but the startup community definitely has less than average number of people interested in football and sports in general. I don't have data on this, it's the feeling I get as someone doing a startup in sports and talking to other entrepreneurs in the bay area.
As a Microsoft employee, OP has access to the private symbol server. Open up windbg and for a lot of MSFT products you can file a bug and send a dump straight to the team... all with a single command and without leaving the debugger.
Other than that, I really enjoyed the author's writing style, it was a good rant
Not directly relevant to the blog post, but Sentry is an amazing piece of software. We're running it on an m1.small AWS instance along with a bunch of other stuff and it is rock solid.
Using async write I/O wouldn't allow them to test the capabilities of the underlying hardware as writes would return immediately.
Also, sixty threads on such a powerful machine is not really a big deal. Even though 0.06ms service time seems small, it is large enough such that you won't have more than a handful of threads competing for CPU resources (they'd waiting for some I/O completion)
No idea what parent poster said (comment deleted), but if you want to work in the US, you can still try for an H1B visa. Your work experience and achievements can be used to demonstrate specialization in the field and serve as the equivalent of a college Bachelors degree. The quota was exhausted for this year, but assuming you are Australian you can shoot for an E3.
Yes, still the #1 choice in my books, especially if your APIs do not match your models 1:1. Tastypie is cute but gets in your way a little too often for my liking. Django-rest is another contender but I haven't tried that yet