Phil Bradley's weblog

GazoPa

Similar image search. Just had an email to tell me that GazoPa is closing its doors. It was an image search engine that allowed you to upload an image to search for similar images. It was a nice interesting idea and a bit of a twist on searching for images, so it's a shame that it's disappearing.

MyAllSearch

MyAllSearch. This is a nice search engine once you get into it. It's one that allows you to input a search, run it and then click on any other engines to see the results that they can provide you with.

With web search you can choose from Google, Yahoo, Bing, Ask (Jeeves), Yandex, Lycos, Metacrawler, Entireweb and DuckDuckGo.

Blogs gives you Technorati, Google, Blogs.com, Blogged and Icerocket.

Other options for search include downloads, images, lyrics, music, torrents and video.

MyAllSearch does exactly what it sets out to do. Clear and clean, nice crisp interface. The only confusion that I found was with the main screen, which is set up so that it looks as though it only lets you search Yahoo, Ask and Images, but if you ignore that and just jump straight into a search, you're good to go.

Alternion

News Feed. I've been playing around with Alternion, which is a social media news collation tool. The idea behind it is that you can gather all of your social media news feeds together in one place. So that's things like Twitter feeds, YouTube, Flickr and so on. You can then get updates, photos and video from all of your contacts on these social networks in one place, rather than keep going from one to the next. You can also update and comment to favourite services. You can also email and DM in one place.

Now, I should be really keen on this. It's a great idea, and should save me lots of time. And yet... having tried it out, it leaves me pretty cold. There are a number of reasons for this. Firstly, I don't know who these people are. I'm absolutely sure that they're 100% bone fide (else I wouldn't be using them) and the small amount that I've seen elsewhere also reassures me. However, the general principle of giving one resource huge access to my personal data and identity does worry me. Take Flickr, for one example. In order to try it out, I allowed Alternion to have access to my Flickr feed - it needed it to be able to access my photographs and those of my contacts. However, what it also allowed Alternion to do was, if it wanted to, delete all of my photographs and video. Would they want to? No. Do I worry that they would? No. But the fact remains - they *could*. For me, that's a step way too far for my liking. I'm not sure if Alternion has requested that Flickr allows this, or if it's the standard access that Flickr grants other apps, but either way, I'm not happy.

The second concern - or perhaps observation is that Twitter rules supreme. Now, I use Twitter all of the time, and follow a fair number (1,500+) of people. So of course it's going to entirely dominate any time line of news that I have. As a result, material that gets posted from other social media apps gets drowned out. Of course, I can just click to view what's available to me from Facebook or any of the other social media sites (and there are a huge number of options available to you - it's impressed me!) However, if I want to see those results, I can just click on the appropriate tab in my browser and have it all there waiting for me. So Alternion isn't actually helping me  here - at most it's doing no more than I can do already. I'm not that hung up on clicks that I worry about another couple to get the data that I want.

My third concern is that with social media I'm not the same everywhere. When I'm in Flickr - I'm not really Phil the librarian, I'm Phil the photographer and artist. When I'm in Facebook, I'm not really in work mode. Importantly, I don't treat people in the same way either. I'd talk to my friend on Twitter in work mode with a particular 'voice' and use an entirely different one on Facebook. The mere fact that I'm in a different space reinforces the fact that I'm treating them differently. For those of you who have ever tweeted from the 'wrong' account will know exactly what I mean, and will know how uncomfortable that could be.

Fourthly I wasn't really impressed with the speed of the service. I expect Twitter updates every few moments but I'm currently looking at a feed that is 18 minutes old. This just doesn't work for me at all. Other commentators have also mentioned this. If it's not instant, I'm not interested.

Finally, there are plenty of other news curation services out there. I prefer Netvibes, or Pulse.news or Zite for example which are neat, simple and easy to use, though it could be argued with the latter two they're slightly different resources as they link to stories, rather than collate social media feeds.

So, in summary, I think it's a great idea in theory but in practice it doesn't do it for me at all. However, if you use lots of social media sites in fairly equal measure many of my comments wouldn't apply, so please do try it for yourself to see if it's a better fit for you.

Strawberry Jam

This is an interesting resource. Strawberry Jam allows you to see the links that the people you follow on Twitter are posting. You can either see real time links, or popular links for the last 8, 16, or 24 hours. The links are properly displayed with titles and you can see the avatars of the people who tweeted, and can click on those to view the actual tweet and any comments they made in detail. It's a really good idea and brings together important or interesting links that you may have otherwise missed.

However, there are problems with the service. The whole system is based on simply ranking on numbers, so the more people who tweet a link, the higher up it goes. While on the one hand this is a good thing, it also means that it's highly likely that I'll have seen the link anyway, since I have Twitter open all of the time. For those who don't though, it's a good idea. What I would prefer is the ability to make use of the Twitter lists that I've got - which would turn this from a 'quite nice' resource to a 'must use.' Being able to see what UK librarians find important and noteworthy and then compare that to US librarians for example - THAT would be exciting. Or to have the ability to artificially rank certain people higher so that if they tweet a link it's going to rocket to the top of my list. All these things could be done without too much work, so I hope this'll be taken as a nudge in the right direction by the Strawberry Jam people.

However, I find the system as it currently stands almost unusable because of the excrable screen display. It's not just bad, it's criminally bad. Take a look:

It's all over the shop! One link that I've looked at has slipped up the screen and behind the search box. The 'mentions' box covers the beginning of the title, leaving me to guess, and it just really irritates me. I should be able to search as well - there's a nice little box that implies this:

Nice in theory, but it simply doesn't seem to work. I'm guessing it's just  a basic bug, but it's irritating nonetheless.

So - bottom line is that it's a great way to search Twitter data in theory, but in practise, they need to sharpen up their act a little bit. However, my gripes are mainly cosmetic and can be sorted quickly. In the long term this could prove really helpful.

d8taplex

d8taplex. Explore over 50,000 data sets containing over 1 million time series. Data is drawn from Canadian, French, Japanese and UK government sites. This engine is VERY beta, so I'm not really going to review it since a)there's not much to review and b)it's unfair to criticise when there's lots that I am sure is going to be fixed in due course. However, if you're interested in data sets, you might want to take a look at this.

Portico

Our Sites and Their Histories | English Heritage. Explore the histories of the sites in the care of English Heritage. Their 400 sites cover the spectrum of English history, from the dawn of the Neolithic to the close of the Cold War. Here you can find out what they know about them and why they are worth conserving.

Well... yes. Look, I hate to be churlish and all, but I've not had a lot of luck with this one. Searches didn't turn up much and quite often when they did I wasn't getting much that was useful, or the criteria such as 'period' or 'type' had not been set. Slight spelling errors (Kennelworth instead of Kenilworth) resulted in zero hits.

Good for what it does, but very basic and I think I'd get more if I did a site search using something like Google or Bing.

Zanran

Numerical Data Search. Zanran helps you to find ‘semi-structured’ data on the web. This is the numerical data that people have presented as graphs and tables and charts. For example, the data could be a graph in a PDF report, or a table in an Excel spreadsheet, or a barchart shown as an image in an HTML page. This huge amount of information can be difficult to find using conventional search engines, which are focused primarily on finding text rather than graphs, tables and bar charts. Put more simply: Zanran is Google for data.

I've tried this with a few queries - civil war casualties, population london, libraries, and I was pretty pleased with the results. Zanran pulled data from tables in PDF format, tables on webpages and spreadsheets. There was a really useful 'hover over' function that brought up the exact data very quickly and I could zip down a list of results easily and smoothly. The SERPS  provide title, summary based on position of search terms, date and source. Search could be limited to region, date, filetype (6 types) and specific site.

If you need figures and statistics, I would certainly suggest taking a look at this engine. There is of course room for improvement, and it can get confused when tables are used  Let me know what you think of it in the comments!

Related articles

• Zanran: New Search Engine That Unearths Data In Charts, Graphs & Tables (searchengineland.com)

A search engine designed to get it wrong:

WackoSearch. Type in a search and get something totally unrelated. It's a slightly odd concept, and really used to try and bring back serendipitous searching. A roulette wheel of sites would work just as well really, so I'm not convinced that the 'search engine' angle is of any particular value at all.

I'll grant however that the sites that do get turned up look very interesting, if you've got time to burn. 'Headless historical dolls', 'Be a fridge poet', 'Museum of bad album covers' all tempted me to spend time poking around. I was slightly concerned that the same sites were coming up on a regular basis though, so I wonder just how big their database is.

Bottom line is that this is vaguely fun for 5 minutes if you're bored. But it's not really a search engine.

RIP:Viewzi

RIP:Viewzi. A bit of sad news - Viewzi, which I always liked - cracking little search engine that provided over 14 different ways to visually search the web has shut its doors. It hadn't been actively developed since the end of 2008. Another small 'victory' for the big boys of whitebread search I'm afraid.

Findebookee

Ebook Search Engine -Find Everything Ebooks. From the site: "This is a free search engine that provides an option to search for and download various PDF&DOC documents, data sheets etc. We don't host any files. Our crawlers harvested a huge database files through different open Internet resources such as blogs, forums, BBS and others. This database is regularly checked for file validity so now you can search within more than three million of live PDF&DOC&XLS and etc. Today we have 28 000 000 document files in our search database and approximately 100 000 files are added daily."

I tried it with a few searches - "web 2" for example gave me 44,000+ results, but the SERPS were quite sparse, and gave the searcher very little to work on - a title, type of file, date and URL. While I imagine that this search engine turns up some real little gems, I also think they're going to be quite hard to locate. However, if you're really desperate, this is a place to go.