Hacker Newsnew | past | comments | ask | show | jobs | submit | dumbluck's commentslogin

Nice post, but:

Don't self-back-pat.

It read more like marketing for the company than anything of much worth.

Don't use the words "web scale".

It is a meaningless term. How many requests per second and how many guest and user sessions for how long of a time is "web scale"? Does "web scale" just mean Christmas shopping traffic on your crappy e-commerce site that no one visits, does it mean you can survive being top link on HN on a Friday morning, or slashdotted, or DoS attacked by anonymous, or survive a fucking flood and the data center lifts off the ground and enough flexibility and strength in the trunk lines to handle a tsunami?

Don't just throw users at it.

Unless testing is very costly and you need every user's eyes on your untested code as possible, that is just stupid. Look at Tsung, Grinder or JMeter, or the many other ways you could generate load as a first step before you do that.

Don't gloss over the details.

Sure you said you were using Rails 3.2 and postgres and a tad bit about the architecture, but who in the hell doesn't know that you need to load balance, need to put the DB servers on different VMs/servers than the apps. Although- having everything on both and some stuff just not turned on and live is not a bad idea for emergency fault tolerance, and you didn't mention that.


Your caps-lock key appears to be broken.


Oh, you're right- thanks! Just fixed the comment. I'll get the capslock key fixed on Monday.


This was the awesome kind of explanation about what went wrong and what was learned that I wish everyone would do.


I appreciate this sentiment, but it makes me think of "The queen is dead. Long live the queen."

"Visionaries" and "cool people" are in many ways good and bad, flawed and gifted, like you and me. It does not serve any good purpose to hold either Jobs or Elon up to the spotlight as if they are any sort of savior.

If anything good comes in the next 10-20 years, it will be because of not only Elon, but people like the rest of us that choose to do something, not wait on others to do something great for us. That said, there is nothing wrong (to say the least) with someone who can drum up the troops.


This is a good attempt, however way, way too glossed-over.

First problem is that nothing is said about Node or Sinatra in comparison.

Second problem is that no mention is made of the significant portion of all gems in rubygems (43,762 at last count) that are Rails-specific. Node is not there yet, and Sinatra can't use a good percentage of those.

Node is about JS. Sinatra is about lightweight/speed. Rails is about features.


That's true. I could easily write a full-length book on this topic. This article is actually meant for the people not using Rails and mostly reminding them that the invisible security stuff is important ;-)


I really wanted to like this, but I think that colors preferences are both subjective and born of habit/experience which leads to wiring our brains to expect certain colors for certain things, because I don't like pastels, and I don't like red as a non-alert color.

It's a good idea, with a lot of thought put into it. The author should keep trying. Maybe there is more than one right option.

Personally, I think the best way to do this is by a "hot or not" site for fonts and colors. Have a page for each language with "hot or not" of random fonts and color combinations until the stats show which combinations certain groups like. Tie in other surveys as well and give away gift certificates to Amazon each month as a prize for over 100 votes that aren't outliers.


Will certainly keep trying. I would say there is no right option and at the same time all options are right.

There does seem something common about themes that appeal to many people but ultimately each has their own correct version of how things should be. Hence my next project https://github.com/chriskempson/base16 Interesting idea by the way!


Agreed. I have a few projects now I've not had time to integrate patches for and they've waited months. I know they are using the patches so they aren't waiting on me, and no one else has complained so it isn't on my priority list. either. In the meantime, I'm submitting pull requests to github projects that I am getting paid at work to work on, some accepted, some denied.


Good post, even though had a bad experience.

I think whether site or startup, it is just a (hopefully informed) crapshoot. In the end, the goals are to be happy, make money, and make a difference. Whether you do that via startup, or other means, doesn't matter. And, if you fail, you (hopefully) learn and can do better next time.


The title is misleading. They've raised enough money that with NY's matching funds they could perhaps pay for the property and preserve it as a historic site, but it won't be a museum anymore than my bathroom is a YMCA. They need millions more.


When you give, your intent is not to pay for a bunch of mailers to be sent out, killing trees and increasing the world's CO2, and your intent is not to hire some Ivy+ grad with a major in making themselves feel better about helping people as they sit at their desk and drink their Starbucks mocha. You want to free slaves, feed the homeless, feed the starving children, cure cancer. Don't give to those that waste that generosity, and don't support sites that don't tell you where your money is really going.

In this case, supposedly the overhead is 16%. That isn't great, but it is in the "meh" category for me. I'd rather give to the Salvation Army that only takes ~5% overhead. In addition, I'd like to see what the 86% going to programs and services is really accomplishing.


Good post, Reg! I think you made Steve's day, and it is good advice for everyone on HN as well. I would be in favor of allowing people on HN to downvote overly negative comments. That would help I think. Not allowing everyone to downvote is a problem. HN karma != helpfulness.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: