notes.husk.org. scribblings by Paul Mison.

Flickr Catchup: A Technique

I’ve been on holiday for a fortnight, so I haven’t had a chance to check Flickr. How to catch up?

Well, the search box on the “photos from your contacts” page has a handy filter: “from your contacts”. Put in space, press return, and you get… an error. Then go to advanced search, change “date taken” to “posted”, and put in the date you went on holiday, and you’ll end up at a page like this.

(Note the date in the URL; you can change that if you want to skip the navigation outlined above.)

That’s sorted by “interestingness”, so you’d hope the better / more noticed photos would come to the top. If you’re more egalitarian, you could always go for “recent”, and then start at the end of the list and work forward. 

If I was away from a computer more often, I might use the API to turn this into a service, but I’m not, so I won’t. Also, as I’ve noted before, it’s a shame more sites don’t offer filtered search to your contacts. I kept up with Twitter pretty well, but I have no idea what I’ve missed on delicious, for example. Hurrah, then, for Flickr.

Foundem vs Google

Foundem, “the world’s most advanced vertical search engine” (as they sell themselves), is one third of an antitrust complaint against Google. They’ve previously complained about being dropped from Google search results, although that’s apparently been resolved since.

In the business news segment on Today, the section that caught my ear was a company representative complaining about Google’s inclusion of video and shopping results (aka “universal search”), specifically mentioning a search for the Toshiba TG01, and the fact that the videos were hosted on YouTube:

(I’ve highlighted the parts of the page that Foundem seem to be arguing against, but removed the Google header.)

Now, I can’t help but agree that, for this sort of search, Google looks cluttered, and you can’t argue that the video isn’t on a Google service. However, while  one of the four “shopping results” links goes to Google’s product search, the other three link directly out to retailers.

So, how does Foundem do?

At least with Google’s results I have a good idea of what the product actually is (a mobile phone); Foundem’s results don’t even list the right category. From Google’s page, I can see reviews and the manufacturer’s site; from Foundem, nothing, until I drill down again. Remember, this is a search that was suggested by the company themselves.

Perhaps instead of complaints to the EU, Foundem would be more successful if they fixed their search engine to, you know, actually work?

RWW vs Facebook; users vs devs

Yesterday, ReadWriteWeb posted a vaguely interesting story about Facebook integrating with AIM, and were then shocked to find the comments overrun with people complaining about how the design had changed and they couldn’t log in. After two and a half pages of comments, Jolie O’Dell, a site admin posted this comment:

We’ve determined by looking at our traffic stats that people are doing Google searches for “facebook login” and coming upon RWW. They see the FB Connect button and assume that RWW is the “new Facebook.”
Sigh.
The Internet Is Hard.

It’s tempting just to post this with a “hah, look at the lusers” comment, but to their credit, the same admin started a reflective open thread:

How can we balance making the Web simple enough for all users while still creating tech cool enough to satisfy geeks like us? And who says either group - nerds or users - is “normal,” anyway?

Rather than answering that directly, I’d like to pull apart some of the differences between “user” and “dev” thinking that led to this happening.

Search vs bookmarks

Devs will usually have access to URLs, whether it be from their memory, from autocomplete (either of history (possibly via a Top Sites / Speed Dial page) or bookmarks). Users seem more likely to search, especially since browsers interpret things in the location bar as searches unless you’re careful.

Search engine sectioning

Most devs would skip the “news” section of the search results unless they were looking for news. Some users seem to click the first link, regardless of sectioning. After all, the link at the top worked before, and does again now. (I wonder if that applies even to ads?)

Doors vs walls

In my experience, devs are more likely to just try to access a page; if it then asks them to log in, they will. If there’s a site they use frequently, they’ll make sure they find and check the “remember me” box, ignoring the “shared computer” warnings; who’d use a computer someone else has?

In contrast, it seems that at least some users explicitly want to find the login page, rather than just waiting for the prompt. Perhaps that’s because they do regularly use shared computers; perhaps it’s because they switch accounts regularly (to see friends photos, perhaps?)

Want more evidence? Look at this list of Google autocomplete suggestions.

Reading site design

As mentioned in the RWW follow-up, ”Users don’t read your copy or look at your branding”. Devs might, but even then, RWW don’t have their site name in the <title>. However, they do have a Facebook-branded “connect” button, which is what the users were looking for (and, evidently found).

Numbers

As a comment on RWW by “a Facebook user” noted, “50% of Facebook users log in to the site daily. There are 450 million Facebook users total. 200 or so people Googled ’Facebook login,’ managed to click on a a news story, landed on this blog, and found the Facebook connect button and posted a comment”. For RWW, and for most dev-focussed sites, that amounts to a huge burst of traffic. For Facebook, it’s almost insignificant overspill.

In Summary

It’s easy to be cynical or dismissive; “look at those fools leaving comments”. Stepping back and considering how much difference there can be in how others view the web might be worthwhile, though.

Wireless

Going on the reaction to the staff post announcing Tumblr Wire, I’m the only person who took an instant dislike to it. It’s the worst of Radar, back. Where’s Recent / Popular / Upcoming gone? Bah. I demand a refund. (Yes, I know Tumblr’s free. That’s called irony. Or sarcasm. Or something.)

Having spent a little more time with it: I still hate it. That scrolling box has zero refindability. Good luck if you want to actually capture anything from it. You thought Twitter had the memory span of a goldfish? Man, that’s like writing in stone compared to this.

On the other hand, maybe I’m just not Tumblr’s desired audience. I write posts of more than two paragraphs; perhaps I should just give up and shuffle off to Posterous. Or ignore all the social crap beyond the Dashboard. At least then I’d only have one source of pseudo-meaningful bollocks.

Edit: Oh dear, it’s even worse. Links from the Wire go through the Digg-bar style tumblupon UI, which has nasty framesets stopping you from easily sharing stuff. Nasty.

Twitter and friend-only searches

Roo Reynolds posted a pair of photos (1, 2) to highlight a feature he says Twitter needs: searching the updates only of people you follow.*

I don’t disagree, but Twitter might find implementing followed-only search difficult. This is because Summize (the acquired company that provides their search functionality) is somewhat disconnected from Twitter proper: it still has own domain (which stayed up on Thursday), for example. Incidentally, this disjoint nature means that people who run private (hi!) aren’t indexed at all.

It’s also why deleted posts aren’t deleted on search, but then another failing actually saves you: the search index is only goes back a few weeks.

Probably the only practical solution in the short term, given I doubt Twitter have the time or inclination to fix these issues, is to fix it yourself. Simon Willison has started running his own local DB, largely for search purposes. It’s tempting to do the same.

* I commented over there, but thought it was worth pulling out for a larger audience.