Follow Slashdot blog updates by subscribing to our blog RSS feed

 



Forgot your password?
typodupeerror

Comment Re:I gave up on SO (Score 2) 618

You have a point.

The StackExchange sites have a weak spot for late answers. The voting and sorting system reward mediocre answers that are posted early over great answers that are posted months or years later. That means that the best answer is sometimes half way down the page and may never reach the top.

It is often problematic that the person who asked the question gets sole control over which answer is at the top via the green check mark that "accepts" the answer. I've seen them choose some really bone-headed answers as accepted on occasion. There is just no way for the community to over-ride them, even with at 10:1 ratio of votes on some other answer.

My other pet-peeve is the large number of separate StackExchange sites with somewhat overlapping topics. It is almost impossible to figure out where to post a question sometimes. Most of the sites have non-obvious rules about what is off-topic. You are likely to ask in the wrong place and get your question closed the way it is set up. For example if you have a question about the security of Google Analytics for your WordPress website running on IIS. You might ask it on Security, WordPress, Webmasters, WebApps, or Server Fault. Most people seem to just ask it on StackOverflow because it is the one they know.

Comment Re:Good (Score 1) 57

It gets much more complicated once there is a load balancer involved. I end up redirecting the acme-challenge directory to a subdomain that gets hosted without a load balancer, generating the certificate there, and then having scripts push it to the load balancer.

The other problem I have is that certbot is not idempotent. Certbot doesn't check if the deploy scripts actually succeed or not, it just assumes they did. If they didn't, they will never get called again. Just running certbot auto-renew is not enough. You have to compare locally available cert to the live installed cert to know if a deploy is needed.

With all those extra check, it works, but it is several hundred lines of scripts.

Comment This really sucks for StartSSL customers (Score 1) 57

This really sucks for customers of StartCom (StartSSL):

  • Your website suddenly stops working with no warning.
  • There is no equivalent alternative to StartSSL

Basically Google (and to a lesser extent Firefox) have handled this really badly. I found out about this issue when I got a new certificate and it wouldn't work: StartSSL certificate gives SEC_ERROR_REVOKED_CERTIFICATE in Firefox and ERR_CERT_AUTHORITY_INVALID in Chrome

  • The browser error messages are cryptic and inconsistent. None of them say what the problem actually is. None of them offer links to the blog posts or bugs announcing the revocation. The only way to figure out the issue is through searching.
  • Google is killing existing certificates without making any attempt to contact webmasters. Google should be putting alerts in Google Search Console for every site that will be brought down by this change. At least Firefox limited the scope such that all existing certificates were grandfathered in.

StartSSL was the only certificate authority at its price point. You didn't have to pay by the certificate. You didn't have to pay for the automated process by which you validated ownership of domains. You only paid for validations of who you are and who your company is. Once you were validated, you could issue as many certificates as you wanted for any domains you own. For a flat fee of $200 per year, I could get all the certificates I needed.

The only alternative that I have been able to find is LetsEncrypt. While it is completely free it has some major disadvantages:

  • LetsEncrypt doesn't offer wildcard certificates. I have a domain with about 60 subdomains. The lack of wildcard really hurts for me here.
  • LetsEncrypt only offers the most basic level 1 certificates. They only validate that you have control over your domain. They don't offer level 2 that validates who you are. They don't offer level 3 that validates who your company is. They don't offer the level 4 extended company validations that give the green bar in browsers.

Comment I miss Firefox in this regard (Score 0) 102

Firefox bookmarks sync is much better than Chrome bookmarks sync. Firefox stored your bookmarks locally and updated them periodically from the cloud. Chrome appears to have to download everything when I start the browser. I get a blank bookmarks bar for a few seconds when the internet is slow and I open Chrome. This is one place where Firefox got the design right and Chrome has it wrong.

Comment As soon as the automated tests pass (Score 4, Interesting) 182

Push to production as soon as the (many) automated tests that you have pass. This means you should have comprehensive unit tests and tests that run in the browser, probably written in Selenium. You'll also want to script your release so that you can do it with the push of a button. Once the tests pass, and the mechanics of a release are trivial, there is little reason to hold up a release.

I worked for a top 500 website (East coast) for 7 years that did weekly releases. Since I left, they decided that wasn't fast enough and now release multiple times per week. I'm now self-employed on my own website and release within an hour of finishing development of a feature.

I started my development career writing firmware for laser printers. When you are shipping code on a physical product, the cost of bugs can be quite high. Especially when it leads to returns or recalls because customers are not satisfied. Our release cycles there were 6 months+. Quite appropriately, IMO.

On the web, the cost of bugs is much lower. In most cases it is the only cost of another release. Sometimes it could cost more because of downtime, but good automated test coverage mitigates that risk pretty well (especially if there is load testing involved). The worst case would be data-corruption, but I've never actually seen that in practice from a release, that has only been related to hardware failure or accidents in my experience.

Comment Re:Real name policy to blame? (Score 2) 456

Facebook has a real name policy as well. It hasn't hindered their growth. The problem is that Google+ has a real name policy, but doesn't require mutual friendship. This leads to a duplicate one way friendship problem.

Here is the use case: you want to add a friend who isn't on the network but you have their email address.
Facebook: You add the user by email. It goes to "friendship requested" status.
Google: You add the user by email. That email address is added to your circles
Then later, the user signs up for the social network, but not using the email address you supplied then friends you.
Facebook: You are friends!
Google: You are friends, plus you have a zombie email address friend in your circles. FAIL!

That and Google+ is full of bugs. For example you open a Google+ account at your own email address. Then you sign up for gmail. This changes the email address of your Google account to your new gmail address with NO WAY TO CHANGE IT BACK. The people in your circles are associated with your old email address. Google has DELETED all the friends from your circles. You then have to re-add all of them.

Comment Re:Aha! (Score 1) 120

The IT department here used on of those "perpetual motion" drinking birds to test the video conference system. A week before the big meeting, they set up the link between our Boston office and our London office, put a drinking bird in front of the camera, and made sure that the connection remained stable enough that it wasn't going to drop during the three hours that we really needed it.
Google

Gaia Project Agrees To Google Cease and Desist 323

Dreben writes "Gaia, an opensource project to develop a 3D API to Google Earth, has decided to comply with a request from Google. The search giant's Chief Technologist, Michael Jones, contacted the project with a request to cease and desist from all past, present and future development of the Gaia project. Amongst other things, they cited 'improper usage of licensed data,' which Google licenses from assorted third party vendors. They are going so far as to request anyone who has ever downloaded any aspect of Gaia to purge all related files. From the post to the freegis-l mail list: 'We understand and respect Google's position on the case, so we've removed all downloads from this page and we ask everybody who have ever downloaded gaia 0.1.0 and prior versions to delete all files concerned with the project, which include source code, binary files and image cache (~/.gaia).' How does such a request, likely to have turned into a demand, affect fair usage? While the API is intended to interface with the the Google Earth service, Google Earth is nothing without the data. Yet at the same time, Google openly publishes their own API which uses the same data in the same manner."
Microsoft

Submission + - Microsoft hands over technical data to EU

hankwang writes: Reuters reports that Microsoft has handed over technical documents to the EU in order to enable the competition to make interoperable software. So far, the EU has imposed fines of €497 and €280 onto Microsoft for abuse of its monopoly. The deadline for this documentation was today. According to Microsoft, the documentation is over 8500 pages.

Slashdot Top Deals

The closest to perfection a person ever comes is when he fills out a job application form. -- Stanley J. Randall

Working...