Dear Yahoo! (or, Dear vBulletin) ... please do something about the Yahoo YUI API!

dalecosp

Just loaded a vB page.

Firebug reports it took 1 minute, 40 seconds, to load
http://yui.yahooapis.com/combo?2.9.0/build/yuiloader-dom-event/yuiloader-dom-event.js&2.9.0/build/connection/connection-min.js

I don't suppose you've got any interest in using something that will actually LOAD when called?

Do you need to buy some hosting? 😛

I hear Amazon claims to have a scalable solution. Or you could just dump the RHEL and go back to FreeBSD ... I have a feeling it would serve more data a tad more quickly.

Sheesh.

</rant>

sneakyimp

YUI! Haven't seen that in ages. I for one am extremely leery of using yahoo/google/whoever for hosting JS files. Afraid they might change the file and break my stuff and also paranoid about how they track folks.

dalecosp

sneakyimp;11050933 wrote:
YUI! Haven't seen that in ages.

Press "CTL-U" on this page 😉

vBulletin uses it. What I don't understand is why it seems to be Just Fine(tm) here at PHPBuilder but when I load one of our sites, at times, it takes forever and a day to load it from the Yahoo network(s)...

I'm thinking to open a ticket at vB, but I doubt there'd be any traction.

As for breakage, that's vBulletin's problem. If they break it too hard I'll just fix it or turn that widget off for the time being.

As for tracking ... I hear ya, but ... well, it's the way things are going. I kind of figure if it was overly terrible, we might've heard from some activist by not. OTOH, if we don't do some checking on such, who's gonna be the next activist? 😉

Bonesnap

sneakyimp;11050933 wrote:
...and also paranoid about how they track folks.

I take it you don't use Google to do any searching online? I have been using DuckDuckGo for the last couple months. It's not bad, and it has some handy features for developers like displaying StackOverflow links at the top that will take you to the page and highlight the accepted answer. It also has some built in custom searches like !wiki [term] which will search Wikipedia.

dalecosp

Bonesnap;11050951 wrote:
I take it you don't use Google to do any searching online? I have been using DuckDuckGo for the last couple months. It's not bad, and it has some handy features for developers like displaying StackOverflow links at the top that will take you to the page and highlight the accepted answer. It also has some built in custom searches like !wiki [term] which will search Wikipedia.

I've got DDG set on FFox, but still let Chrome use its mama for searching. Not sure about DDG ... I like their philosophy, the result set seems apropos and all, but a few things bother me because, I assume, I'm Just Too Used To Google. One of the sillier ones, I think, is that I don't like endless scrolling and apparently prefer paginated SERPs...

Oh, and the bang that probably gets the most use for me at DDG just serves to keep me tied to Google ... !gi [something] ...

sneakyimp

Just because you use google search doesn't mean you have to let every jackhole under the sun track you. I use google to search and this doesn't bother me especially. I do, however, add certain google (and numerous other) tracking domains to my hosts file so I never visit them. It bothers me that I cannot edit the hosts file on my phone in the same way. I prohibit 3rd-party cookies. Every chance I get, I try to avoid giving up my email address or personal information. I truly believe that doing so tends to reduce the likelihood that I will suffer identity theft or that I will be a victim of social engineering by some hacker or criminal.

Bonesnap

sneakyimp;11050967 wrote:
Just because you use google search doesn't mean you have to let every jackhole under the sun track you. I use google to search and this doesn't bother me especially. I do, however, add certain google (and numerous other) tracking domains to my hosts file so I never visit them. It bothers me that I cannot edit the hosts file on my phone in the same way. I prohibit 3rd-party cookies. Every chance I get, I try to avoid giving up my email address or personal information. I truly believe that doing so tends to reduce the likelihood that I will suffer identity theft or that I will be a victim of social engineering by some hacker or criminal.

I would wager that most of Google's tracking is done via their Google Accounts. But I think including a JS library via a CDN is probably on the lower end of the scale of clandestine things a corporation can do regarding tracking. I do agree though of course that it's just better to host your own file for the other reason you mentioned.

sneakyimp

Bonesnap;11051011 wrote:
I would wager that most of Google's tracking is done via their Google Accounts. But I think including a JS library via a CDN is probably on the lower end of the scale of clandestine things a corporation can do regarding tracking. I do agree though of course that it's just better to host your own file for the other reason you mentioned.

I think I respectfully disagree. The apache log alone for these google-hosted and yahoo-hosted libraries can provide a treasure trove of detail about what sites you visit which can build a far more detailed of your profile than is possible from just a google account. Having access to your email is certainly invasive enough, but knowing what websites you visit outside of their account system is arguably even more invasive.

dalecosp

sneakyimp;11050967 wrote:
It bothers me that I cannot edit the hosts file on my phone in the same way.

Hmm. I can. #got_root ?

Derokorian

dalecosp;11051089 wrote:
Hmm. I can. #got_root ?

Android FTW... I can totally do this too! You can also set up a proxy through a private server, and then have your server set up to not pass through requests to any domains you don't want to hit. I've done this with iOS products, but never gone through the effort with Android (because well, I can edit hosts on my Android).

sneakyimp

Derokorian;11051091 wrote:
Android FTW... I can totally do this too! You can also set up a proxy through a private server, and then have your server set up to not pass through requests to any domains you don't want to hit. I've done this with iOS products, but never gone through the effort with Android (because well, I can edit hosts on my Android).

My first smartphone was a Nexus One I think. While I like that it was much less of a 'walled garden' type philosophy behind it (e.g., I could just drag audio files in a folder and they'd be available to play, for instance) the operation of the phone itself left a LOT to be desired IMHO. I switched to iPhone and have been mostly happy.

I have no interest in hacking my phone to gain root access. That proxy suggestion sounds interesting, but wouldn't it introduce a lot of latency in your phone's operation?

Derokorian

sneakyimp;11051101 wrote:
My first smartphone was a Nexus One I think. While I like that it was much less of a 'walled garden' type philosophy behind it (e.g., I could just drag audio files in a folder and they'd be available to play, for instance) the operation of the phone itself left a LOT to be desired IMHO. I switched to iPhone and have been mostly happy.

Gross ;-) (just playing, not trying to start a flame war between iOS and Android)

sneakyimp;11051101 wrote:
I have no interest in hacking my phone to gain root access.

I don't remember hacking anything... Its more Apropos to installing a specific version of grub and configuring it on my Linux box. Jail breaking OTOH, is very much hacking and not something the hardware and software is designed to support.

sneakyimp;11051101 wrote:
That proxy suggestion sounds interesting, but wouldn't it introduce a lot of latency in your phone's operation?

I'm sure it does, but the trade off is not loading gifs and webms over 4G. The latency of the proxy seems more offset by not loading a bunch of random ish I don't want. I can tell you once I started doing the proxy, my tendency to get close to my monthly bandwidth dropped off.

dalecosp

Derokorian;11051113 wrote:
I'm sure it does, but the trade off is not loading gifs and webms over 4G. The latency of the proxy seems more offset by not loading a bunch of random ish I don't want. I can tell you once I started doing the proxy, my tendency to get close to my monthly bandwidth dropped off.

I'm not at all sure it does, (although I'm agreeing with the rest of this, as long as we mean a 'caching proxy') as ISP's and big corporate nets have used caching proxies for years to alleviate slow page-loads, etc. (it also works pretty well to catch people who use forums at work, d'load warez and pr0n, etc. )

Bonesnap

sneakyimp;11051073 wrote:
I think I respectfully disagree. The apache log alone for these google-hosted and yahoo-hosted libraries can provide a treasure trove of detail about what sites you visit which can build a far more detailed of your profile than is possible from just a google account. Having access to your email is certainly invasive enough, but knowing what websites you visit outside of their account system is arguably even more invasive.

I think you may be underestimating what is under the Google Ecosystem (TM). While visited sites can reveal information about a person, most people don't really "surf the web" like the media and most people think. People tend to be habitual creatures and if you really look at it only visit a handful of websites on a daily basis.

The Google account on the other hand extends waaaay beyond just Gmail. But email to me is the most personal thing about you on the Internet. My email account dates back about a decade, and there's a decade's worth of conversation history in there. Friends, families, ex-girlfriends, reservations, file attachments, a plethora of other websites I have signed up with and sent their "welcome" email, airline tickets, you name it, it's probably in people's email account. I think that sort of thing builds a far larger profile of a human being than a list of websites. And that's just Gmail. Don't forget Google's mothership: the actual search. I believe Google is allowed to keep your info up to 18 months. That's a pretty decently long history of Google searches. Then there's Youtube, Google+, Google Maps, Google Translate, Google Latitude (this one, if enabled, will literally track your movement via GPS. I have spooky stories about this one), Google Drive, Google Play, Picasa, Blogger, Google Wallet, Google Calendar, Google Webmaster Tools, Google Analytics, etc. I mean man... look at all that stuff, and I barely scratched the surface of Google. And it's all under a single Google Account (TM). They amalgamated their services some time ago under one umbrella. And hell, they want you to log into your Chrome browser too. It's like you're logging into the Internet! Not to even mention integration with your Android device. I can tell you about all sorts of spooky behaviour because of the Google Ecosystem.

Yes, if you avoid the Google ecosystem you can avoid a lot of that stuff, and a lot of people do. Some people make it their mission to never use a Google product. Others do the same but with Microsoft, Yahoo, or Apple.

Tracking cookies is one thing, but I hardly see how including something like...

<script src="https://ajax.googleapis.com/ajax/libs/jquery/2.1.4/jquery.min.js"></script>

...suddenly opens the floodgates, unless you mean that creates tracking cookies on one's computer? Is that even possible?

sneakyimp

Bonesnap;11051143 wrote:
I think you may be underestimating what is under the Google Ecosystem (TM)...

You make good points here.

Bonesnap;11051143 wrote:
Tracking cookies is one thing, but I hardly see how including something like...
<script src="https://ajax.googleapis.com/ajax/libs/jquery/2.1.4/jquery.min.js"></script>
...suddenly opens the floodgates, unless you mean that creates tracking cookies on one's computer? Is that even possible?

When I use gmail or some other google service (google drive, etc.), at least I'm consciously aware that I'm doing so. When I visit myguiltypleasuresite.com or something, it's a bummer to think that there might be tracking code on that site that further accumulates data about me without me actually thinking about it. I think it's discourteous to your users to include this sort of thing generally speaking. Personally, I try to add such tracking type projects to my hosts file as much as I can.

Derokorian

Bonesnap;11051143 wrote:
People tend to be habitual creatures and if you really look at it only visit a handful of websites on a daily basis.

Agreed. Myself, even when looking for new content I tend to search on the G and find a site I know of to read on. I find most of my friends and family are the same. They may consume lots of information - they tend to consume it from familiar and trusted sources.

Bonesnap;11051143 wrote:
Tracking cookies is one thing, but I hardly see how including something like...
<script src="https://ajax.googleapis.com/ajax/libs/jquery/2.1.4/jquery.min.js"></script>
...suddenly opens the floodgates, unless you mean that creates tracking cookies on one's computer? Is that even possible?

Yes its entirely possible to send a cookie back in any http request. Its also entirely possible to know what site you visited when you downloaded said resource, if your browser sends the referrer header. OTOH, browsers cache resources - so the more sites that use the centralized jQuery the less times someone will have to download it to their device. So they may know you visited example.com but not the next 10 sites who use their hosted resource, because now its cached on their device. IMHO, I think every one should try to use the hosted resources whenever possible as I believe it will lead to a faster internet for the average user (but I don't even do it myself...).

One big thing I think you guys are also missing, G+, FB, Twitter, etc buttons on websites. These also give cookies and can track where you got them from because they load JavaScript into the page. I have no idea what the js they load does. One thing someone told me the other day is that FB knew my last 3 jobs, and where I've lived. However, I haven't updated FB in over 6 years - how did they get this information? I literally didn't give it to them willingly...

laserlight

Derokorian wrote:
One thing someone told me the other day is that FB knew my last 3 jobs, and where I've lived. However, I haven't updated FB in over 6 years - how did they get this information? I literally didn't give it to them willingly...

Did you or your employers place your information on the public Internet? If so, then perhaps anyone with sufficient data to identify you can find out your last 3 jobs, and since you have provided sufficient data to Facebook, Facebook can find out your last 3 jobs, even if they have not done so yet.

Another possibility is that former colleagues have been identifying you as such on Facebook. You may have ignored these, but despite your lack of confirmation, there's no smoke without fire, so Facebook can reasonably suppose that you were indeed a former colleague, especially if more than one such former colleague rats on you to Facebook.

dalecosp

This is all Big Data type stuff, and it's reality everywhere. So many pages are "mash ups" with so many JS widgets on them, and yes, they do track you anywhere that has such widgets (and Google would be one of the biggest doers of such). I guess everybody knows about Target notifying the father about his 13 y.o. daughter's pregnancy? All the big *.com's do this stuff. Use Google for searching, throw in FB and LinkedIn, and there's a ton of data about every one of us out there, plus predictive analytics and lots of back-looking analysis of stuff so they can target advertising to you.

And then there's the NSA, which has agreements with some companies that have been divulged, and probably has them with the rest that haven't been. It's pretty hard to get "off the grid" if you make your living by working ON it. 😉