Google Blogoscoped

Tuesday, July 27, 2004

Browser Buttons For the Future

The browser interface (and web page metaphor) has been stagnating since the mid-nineties. The web and its links are basis upon which browser creators put the features of following up on those links, the back and forward buttons, a print button, and other basic functionality.

But there are many things left which should in the near future be handled by browsers to enhance the browsing experience; to in fact change the way we communicate using the web.

They all come in the shape or innocent buttons. But they hold a lot of power.

“Comment”

Just like every monitor “supports” post-it notes, a comment button should be integral part of any browser. Once you press it you select the part of the web page you want to comment on. Furthermore you select the reader group you want to share the comment with. This could be public, or a predefined “co-workers” group, or a predefined “friends” group.

Once your comment is posted you have the chance to edit it later on. Potential readers can be alerted of comments.

What it takes to implement this is both an annotation server, or more likely several ones – like there are different chat networks forming separate “dimensions” of IRC.

“Add Feed”

Today when you click on RSS feeds or Atom feeds, you are redirected to a strange page including headlines. Of course these XML files are in machine-readable form. They are not intended to be read by humans, so why are they displayed in the browser right where HTML pages are displayed?

Future browsers should all ship with RSS readers (only this thing would not be called RSS, but something like “news feed”... there are only so many abbreviations a non-geek should digest). These readers would be included in the main interface – possibly a sidebar to the left. Clicking on an RSS or Atom feed should simply open a pop-up: “Do you want to subscribe to this feed? (Yes | No)”.
Once subscribed, feeds are available to the left-hand subscription list.

Additionally, future browsers should be smart enough to understand the meta-section of an HTML page (or one of its links) which points to the news feed. Clicking on an “Add Feed” button would automatically scan the page for a subscription option, and if available, add the feed to the feed list.

“Save”

Why can’t I just type over the text of a certain page (if the webmaster allows me to) and click on a “Save” button?

Wikis today offer the possibility to search for the “Edit this page” button, click on it, upon which one will switch into an edit mode.

This edit mode of Wikis is nice enough and the best we have today; but this shouldn’t limit us in the future. What we need is the ability – just like we take for granted in non-HTML formats like Word documents – to mark the text and type over it.

All it would take for this is for the server to send a message to the browser. This is part of a technical protocol. This should not be part of the human interaction. All the reader needs to become a writer is a little icon somewhere indicating that this page can be edited.

“Provide Password”

Clearly there should be a way for the user to provide a site with a password without having to think of one. I don’t know the technical details necessary to solve this, but find it ridiculous I have to enter different passwords into every other site (and manage storing them somehow – at least here, most browsers provide good help).
If necessary, I’m ready provide the site with a fingerprint, or let it scan my eye.

“Show Links”

I can see outgoing links from any web page. They are mostly blue and underlined, or otherwise separated from the rest of the text.

But why is it I can’t see the incoming links for any web page?

I can go to Technorati or similar services. This is an extra-step, an extra-web site to go to.

Every link builds a two-way bridge; it’s an indication of a connection, a relationship. And it’s important information. So why can’t I see the incoming links if I want to?
I think a future web browser should have a “Show Links” button.

Possibly a variant of this button could be a “Show Citations” button. You would mark a sentence or paragraph, click the button, and instantly see titled links to pages quoting this text.

“Page Back”

When I grab a book in the library I get instant information on its overall size. When I open an individual page I get the context of a previous page, and the next page. I can go to these pages without knowing the specific book (I only need to know the general concept of books); I can go to these pages without having been there.

On the web, to go back to the last page I must have been there. In other words I can only go back in my own browsing history; not in the web site I landed on. But I might want to, especially if I’m reading within a narrative structure.

In fact, there is already an HTML construct to handle this job of letting the author of a web page indicate previous and following pages. Some browsers, like Opera, understand it well and render buttons.

These buttons are integrated into a standard navigation bar on a standard position, which is of course the whole point.

The key to usability is knowing what to expect from the interface, so the interface ideally reuses design when the underlying concept is similar (and already well-established). A “Page Back” and a “Page Forward” button should be integrated into every browser. Added to that, the reader should get the feeling of how many pages there are to come; the user should get an instant feeling of the size of the information provided (if this web site has a narrative structure).

“Blog This”

Of course, a blog-this button would be needed. As things go it would be made optional, and would need to be configured in the settings. But this is certainly more desirable than being restricted to the Windows-only, Blogger.com-only Google Toolbar (as good as it is).

“Add Thumbnails”

In current web browsers, links come in a simple shape. Mostly they are text-only renderings of the page to come. There’s nothing inherently bad with that, but it doesn’t always have to be that way.

A future browser could easily feature an “Add Thumbnails” button. Clicking it would render a small preview image of the site to come next to every link. This would especially enhance the scanning of longer link lists.

“Translate”

Hopping from page to page on the web you will often find content in foreign languages you don’t understand. Today you are forced to pay a visit to a Babelfish web app.

A future browser on the other hand might have a “Translate” button built right into the standard interface. Clicking on it would translate the current page into the language as defined in the preferred language settings.

“Speak”

There are text-to-speech plug-ins, and some of them are really good. But I want my native browser features to be able to speak to me. This must not be a TTS browser per se; just the option to have a text be read to me would be nice once in a while.

“Search Visited”

Did you ever want to go back to a page you saw some weeks ago, but you couldn’t find it anymore? After all bookmarks are often highly inefficient (especially since you might not always know ahead of time which will be the pages you want to revisit). Google and others won’t always help in this situation either.

A browser could integrate two features to make it possible to find older pages that have once been visited.

First, a “Search Visited” button. You could enter any text to find pages within the history.
Maybe some browsers already offer this. But what if you cannot recall any specific words from the text you want to find again?

A “Browse Visited” option would be the solution. You would be offered a thumbnail collection of all pages you visited within the last weeks (or months), and click on the page you were looking for.

Search Results Usability (Video)

Semantic Highlighting [MPEG]: A by now ancient (1998) look at how to optimize search engine results by the use visual indicators (like a pie-chart showing word-frequency within the target document).

What Makes a Blog a Blog?

This week you can see CNN.com’s blogosphere unfold. It’s linked right from the CNN homepage.

There’s also news stories popping up in mainstream media, and sometimes they are met with a certain resistance towards what some might consider YAH (i.e. Yet Another Hype). Didn’t people just slam a new word on something? “Let’s call this ’blog’ and pretend to have reinvented the wheel?” Or is it a kind of elitist slander towards the uninitiated?

In answering those people I’ve been thinking about what makes a blog a blog. In other words, how we define a blog and where we draw the border to the non-blog.

So far I’ve been thinking of news sites – such as CNN.com – as blogs of their own. Any regularly updated web page which doesn’t put emphasis on navigation but delivers content straight-away. (But if CNN.com has to invent their own blog, as they currently do, how could they be a blog themselves?)

Typically a blog must share some of the following characteristics:

I do not believe a blog needs to be a personal diary, like many mainstream media people do – in other words, mumblings only of importance to the person and his friends. (Yes, there are those blogs too, and they also serve a need.)

Technorati in their welcome note for CNN readers, titled Blogging Basics, writes: “A weblog is a website that is updated frequently.”

Updates are important. Of course you are not forced to update a blog, but then it dies. (You would miss out on a great learning and sharing tool at the same time – it’s not without reason some think of their weblog as a backup brain.)

Then again many of the attributes as listed above are mostly technical ones. But when I look at incarnations, like CNN.com’s latest self-ascribed attempt at blogging (with e.g. an RSS feed missing) I get the feeling a blog is when you call it a blog.

Now if this smells funny, let me explain.

Let’s say someone would hand you a fresh blank paper everyday, along with a pencil. Your only job would be to fill this paper with random musings. Whatever comes to your mind, whatever interests you. And now let’s say this would be within a bigger room, with lots of people, all of them doing the same job (filling a blank sheet of paper day in day out), and you could have the chance to walk around and look at other papers.

Now of course the paper of someone else might inspire your own paper for the day. You start talking back in this kind of pull-only, bi-directional form of conversation. And whenever you do, you connect a little red thread from your own paper to the paper of the other person.

You guessed it – the room is the collection of all weblogs (what we call blogosphere) and the individual paper is a blog. The pen in your hand is a simple tool to get your thoughts down on it; the blogging software (or knowledge management tool, or content management system, or whatever you want to call it).

Would I think the same thoughts if no one would hand me a blank paper? No, I wouldn’t. I wouldn’t crack my head everyday at bringing a fresh view on things, or digging out stories. I wouldn’t have become an expert on a chosen topic just because I would expose myself to different views on it everyday. Filling nothing with something flexes your brain muscles.

Do I need to have a certain urge to actually fill the blank paper? Yes, certainly. Not everybody wants to get into the act of a daily weblog. It’s a form of communication which amplifies what you have to say, and makes sure you say it. Before you consciously started a blog you certainly went into blog-like directions prior to taking that step; you certainly do what you love doing (or else, suffer the blogger-burnout people talked about recently).

Because the human interaction defines this blank-paper communication room and its people and creates similar meaning to it as we acknowledge in the blogosphere, we can conclude the social context is what defines, or rather, what makes a blog. The social context which hands you a blank paper everyday, expecting you to fill it. The social context which calls whatever you do a blog.

Gmail Information Architecture

“Gmail revealed to me my email behavior – something I hadn’t previously given much thought. By making certain things easier (and others more difficult), Gmail showed me how “typical” email applications weren’t necessarily designed according to how I used them.”
– Dan Brown, The Information Architecture of Email, July 14, 2004 [Via Amy.]

Google Search Features (Humor)

A bunch of Google web search features (including the “I’m Feeling Sad” button) we’ll probably never see. At least not in the near future.

SearchIRC

SearchIRC searches through the Internet Relay Chat network to find relevant rooms (or to check who’s online).

Googlehouse

Enter the Googlehouse, “an online process that builds a house with images of domestic rooms (living room, tv room...) picked up on the internet using an image search engine."

Bloglines Backlinks

Did you know you can check backlinks to your blog by using Bloglines? You can use the following URL (replace “www.example.com” with your own URL):

http://www.bloglines.com/citations?url=www.example.com

Bliki

What’s a Bliki? It’s a blog with wiki support. According to Wikipedia this means “that after (or before) an article is posted to the blog, it can be edited, either by anyone or by some some group of authorized users."

Search Engine History

Here’s a little Search Engine History by WebProNews.

Another search engine history can be found at this blog.

Google News Visualization

In the News is an interactive visualization of ongoing Google News people and events. Click on any news block to zoom in on the event and see it connected to previous coverage. Creator Michal Migurski also offers the Flash/ ActionScript source, and the raw XML data for others to use.

Webbrain

Webbrain is a visual (Java-powered) cluster search engine. [Via World 2 Come.]

"Revenge of the Sith" Meme Continued

Here is the pagecount update for “Revenge of the Sith”. As you can see Google has a “slow” start (well, two days only) but is now leading the pack:

“Revenge of the Sith” PageCount
 July 25July 26July 27
Google3,8503,86024,400
Yahoo6,65013,20024,100
MSN7108881,158
 
Technorati191266335
Feedster447708830

Things You Can Do With Very Little

“Send spooky emails
(Amusement Potential: 15-60 minutes)
Look up someone’s CV on the web, do some research on them via Google and then send them an email full of personal references claiming to be an ex-work colleague who fell in love with their shoes. Or something.”
– The Geek, When you’re bored..., 19 June, 2004

Other suggestions of what to do when you’re bored include:

More at When you’re bored....

Search Timeline

Resourceshelf points to a Web Search Timeline [PDF].

New Google Domains July

Google Inc registered the following domain names this month:

John Battelle concludes these domain-names hint at the coming of a Google IPO roadshow via webcast.

New Google-domains registered by others than Google Inc include:

Find the full list at
Recent Domain Name Registrations That Include the Word “Google” (July 2004)
Compiled By Gary Price, MLIS

Advertisement

 
Blog  |  Forum     more >> Archive | Feed | Google's blogs | About
Advertisement

 

This site unofficially covers Google™ and more with some rights reserved. Join our forum!