Google Updates Its Face 548
whereiseljefe writes "About 12:00 am Central Time, at least when I saw it, Google changed it's face. Before it was a simplistic search engine, with a minimal front page, and now has become even more so. Those pretty tabs we have become accustomed to are now gone, and in the search results, the "summary" section at the top is now a faded blue bar (see here with a search returning ads). And the ads are a little more low key. Nice to know they are cutting back on their interface rather than adding spastically like Yahoo." Other folks noted that they've added Froogle and Local Directory pages have now been given links on the front page. Which is good, since inclusion in the main page tends to mean ready for prime time.
Froogle is getting a lot better (Score:5, Interesting)
Despite how much I hate advertising, when I actually *want* to see adverts about a product, it is hard to find.
Meanwhile... (Score:5, Interesting)
The same godawful color schemes, ugly nexted tables, awful HTML code, etc.
Maybe slashdot should take a cue from google and update themselves.
This is news? (Score:0, Interesting)
A website added some links! News at 11!
Re:About Face! (Score:5, Interesting)
But look at the poll results: Google only got 3%. I don't think it's overused in a bad way; I think it's overused because you'd be foolish to use anything else!
Comment removed (Score:0, Interesting)
high key ads (Score:5, Interesting)
I like it. (Score:5, Interesting)
I particularly like the idea of seperating "Froogle", I hope in the long term this will bias commercial support away from the generic pages. When I want to know about Hawaii "per se" I am just not interest in tour operators and hotels!
Re:Why are they trying to look like Yahoo!? (Score:2, Interesting)
Definition (Score:5, Interesting)
In firefox... (Score:2, Interesting)
Re:Generally so, but not for /, (Score:4, Interesting)
Google isn't valid HTML either [w3.org]. And they still use an embedded style element rather than a highly-cachable external stylesheet, and still use crap like <body bgcolor=#ffffff...
Fatal Error (Score:5, Interesting)
I don't like it (Score:5, Interesting)
All they did was change the layout.
With the old layout I could navigate the page blindfolded.
I had mouse movements down pat.
The tabs being close to the first search result was handy.
Now you have to navigate to the very top, center of the page.
I've never seen a reason to change an interface, just to change it.
It wasn't broken, why fix it (Score:5, Interesting)
Now that MS and Yahoo are picking up the pace and investing heavily against Google to outcompete it, is this really the time to change Google's look? Search functionality may be all that matters to a geek, but Google is mainstream now and has to worry about mainstream concerns, like "Branding". Google's old look was part of the Google "brand".
I may come off like Chicken Little given that this is such a small thing to be concerned about, but sometimes in the face of heavy competition the smallest things can turn the tide. I've seen it happen.
time will tell... (Score:3, Interesting)
Anyways time will tell how this goes... On the flip side this is one site that can handle the
I want my old google back (Score:5, Interesting)
For instance, Google groups search result pages looks like they are formatted for a 800x600 resolution screen. Viewing it at a higher resolution forces a large white space between the search listings and the ads. I would have much prefered for the results to take up this space, fitting more results on the page at a time. If the group name is long, then the "View Thread" becomes unnatural looking wrapped between two lines. (example [google.com])
Maybe it's just new, but hopefully it'll grow on me.
Re:About Face! (Score:5, Interesting)
Re:Why are they trying to look like Yahoo!? (Score:1, Interesting)
Re:In firefox... (Score:3, Interesting)
Re:Fatal Error (Score:5, Interesting)
I wonder if they skipped the doctype tag because it's relatively pointless for this level of basic HTML, and wasn't worth the bandwith demands to include it.
Re:About Face! (Score:4, Interesting)
Other languages not supported yet (Score:3, Interesting)
I see the translation teams have some work to do...
Re:This is news? (Score:2, Interesting)
I suppose when you are talking about a website that has made SUCH an impact in our lives that people now use "to google" as a verb (expect it being added to dictionaries in a few years time) then it is pretty significant. Come on, we all (ab)use it daily.
Re:Meanwhile... (Score:5, Interesting)
However, on the tables and HTML code side, you're quite right.
Re:Generally so, but not for /, (Score:4, Interesting)
HTML is broken [attrition.org], not google.
LaTeX - it's not just for bootie calls
Re:Slashdotted... (Score:5, Interesting)
More interesting is the wayback machine's caches [archive.org] of Google:
Re:Mirror (Score:5, Interesting)
Re:Froogle is getting a lot better (Score:2, Interesting)
Re:I don't like it (Score:3, Interesting)
I think taking away the pretty tabs was a step backwards, but it's nice to see that they've made the site easier to navigate.
Re:Mirror (Score:1, Interesting)
Anyone else notice the tracking? (Score:5, Interesting)
I saved a copy of the page at the following URL: http://www.phrise.com/google.html [phrise.com] (I added <base href=http://www.google.com/> at the top.)
If you're seeing the same thing, please reply...
Re:About Face! (Score:5, Interesting)
Re:About Face! (Score:5, Interesting)
What I find most ironic is that this is one of the rare occasions when I knew before the article appeared on Slashdot!
In fact, that would suggest that Google is one of the few sites that I visit more often that Slashdot!
Revolution (Score:5, Interesting)
I can. Many people look at Google as an authoritative source. Hence, the gripes you hear about businesses who are made or broken by their Google search result rankings. Now we have Froogle. The danger is worsened even more if people view Froogle as authoritative. Last week I was searching for non-U.S. made baby strollers. I found strollers using normal Google that I couldn't find in Froogle. The only thing I trust Froogle for is to view quick thumbnails of products. For most of my product searches I must rely on complex queries to bypass the senseless froth I see rising to Google's surface more and more these days.
My one wish for Google is for it to face stiff competition. I look forward to anyone who can topple Google with a better engine.
Re:New google fizzles (Score:4, Interesting)
- Search for Gnome [google.com]
- Search for KDE [google.com]
- Search for Enlightenment [google.com]
- Search for Linux Kernel [google.com] has a link from inside kernel.org as the 4th result!
Not one returns the actual home page of these projects! Yikes.
Not only looks -- directory search missing :-( (Score:3, Interesting)
http://www.google.com/search?hl=xx-bork&q=java [google.com] and compare it with the new google interface:
http://www.google.com/search?hl=en&q=java [google.com]
I think this is a sad loss of functionality. The link to directory categories served two purposes for me: First, it was some kind of extra "quality" check -- if a web site was listed in the directory, it was more likely to be the site I was looking for. Second, it informed me in a non-intrusive way that a directory category existed that'd probably help me in my search.
And to add insult to injury, Google has removed the simple link to Directory from the "tabs", so you have to first click "More>>" to find the Directory search, making it even hard to use it. I wonder if this is the first step in stopping to support the dmoz directory?
Re:Meanwhile... (Score:5, Interesting)
It's been done [alistapart.com]
I don't know why it hasn't been implemented yet. I wonder if the /. crew is working on it.
They may also have improved blogger indexing (Score:2, Interesting)
Today I can search them. I wonder if they've done a major maintenance cycle on their indices? That would fit with the speedup reports.
BTW, I do enjoy using the new "define" feature. Try "define: glycoprotein" for example.
The cost of web standards (Score:2, Interesting)
Then I remembered that they get 1800 queries per second, that means also about 1800 homepage/result page views.
So to have the page validate would cost them about
1800 * 3600 * 24 * 150 bytes (size of the DOCTYPE def. to be added) = 21.72 Gigabytes of Bandwidth per day!!!
You can't refuse these kind of savings...
On the other hand, some of the special language versions certainly add more than 150 bytes to the homepage length...
just my
Q.
Re:Generally so, but not for /, (Score:1, Interesting)
But surely using a stylesheet carries extra processing/bandwidth overheads?
No. Firstly, it's true that there are downsides, it's just that in most cases, and especially in the cases of high-traffic dynamic websites, they are vastly outweighed.
Firstly, bandwidth. A second HTTP request will include headers that, in the case of a simple page like Google, will actually be almost as large or possibly larger than the stylesheet.
This is outweighed by the fact that in almost all cases, the stylesheet will be retrieved from a cache rather than the Google servers. Either the browser's cache or the ISP's cache (almost all ISPs have one, and use of interception proxies is growing). This is a very popular website, and the stylesheet doesn't change much, so caches will give the stylesheet priority in the cache, so it will almost always be available from the ISP cache, even if an individual user has never visited Google before.
Also remember that the CSS cannot be cached when it is embedded in the HTML, as Google's pages are almost all inherently uncachable.
Another thing to note is that even when the external stylesheet passes its expiry date, browsers and proxies can revalidate it ("make it up to date again") without downloading it again. All it takes is a single 304 Not Modified response from Google to say "yep, nothing's changed on this end".
Secondly, a second resource, such as an external stylesheet, will often require a second TCP connection. More and more browsers are implementing HTTP 1.1 style persistent connections, and old-style HTTP 1.0 persistent connections are widespread, so this isn't as much of an issue as it used to be. Even so, the fact that people will almost never download the stylesheet from Google's servers means that the impact of this second connection, where it is necessary, is far lower than you might expect.
Remember, as well, that external stylesheets can be offloaded onto a static-only web server, in the same way Slashdot do to serve images. In-kernel webservers like Tux can dramatically speed up serving speed, but that only works for static files like images and stylesheets, not dynamic pages with embedded style elements like Google currently use.
Finally, if they are worried about bandwidth, external stylesheets are very compressible, and it's many orders of magnitude more efficient to compress a single external stylesheet than to compress each and every page view dynamically. Even if they want to compress each page dynamically anyway, it will still reduce resource use by removing the styles from the pages.
Interestingly... (Score:3, Interesting)
Re:About Face! (Score:3, Interesting)
Re:Not new! (Score:5, Interesting)
P.S. I am now seeing at work when I look with IE but not with Opera.
Re:Fatal Error (Score:3, Interesting)
I don't think an ISP cache will interfere with that request, and as somebody else said, when you deal in the kind of traffic google has, every (literal) bit counts.
-Rob
Re:About Face! (Score:2, Interesting)
Waaayyyyy back in the day when I used to work at D.E.C. ('97 to early '99), I and most of my friends and co-workers swore by Altavista... Guess it also didn't hurt that I could feel the hum from the servers that powered it through my chair some days. Boy have things changed, but then, not much changed for the better after the Comwhaq buyout.
One major annoyance with Google lately though... Those stupid results that come back as the 2nd, 3rd, and 4th result for evey search I've run [google.com] in the last month+... You know, the ones that say "Find 'x' using the free 2020 search toolbar", and, "Find 'x' on smartpages.com," and "Find 'x' With Free Websearch Tools." What is up with this??? Why for they cannot make these go away, bitte? If I wanted to search for these somewhere else, I would do so. Why is Google doing free advertising for these people, as they are obviously not paid ads, but standard returns that appear to be just a database/dictionary/meta-tag exploit... Someone at Google must know about this, and I just don't understand why it's been allowed to continue, as this has completely ruined so much of their credibility since now 3 out of the first five results of any query are now completely unrelated and inaccurate =(
Otherwise, I've been loving on Google about 2 1/2 years now, and I haven't really looked back, even when I read sites (1 [salon.com], and 2 [google-watch.org]) that called into question google's privacy practices, I wasn't really deterred, but these bad returns may be all it takes to make me start considering another move.
Re:Fake hits (Score:1, Interesting)
When I was faced with a link farm, I reported it and in a week or two the search results were much improved (not sure how long it took, I didn't search for it again for a while). They do react, when you tell them about it.
Economics (Score:2, Interesting)