Even a URL tells a story | Hacker News Reader

Even a URL tells a story

reillybrennan.com

75 pointsarepb43 comments

This story is pretty short: the Sony page is an Enterprise Java (TM) shopping cart, not a brochureware page in a CMS.

Apple also is a heavy user of Enterprise Java. Try finding pages where they can take input.

That said, yeah, given the choice I would a) devote engineering resources to getting human-readable URLs (SEO and UX benefits for reasonably little work) and b) prefer frameworks which make that the default because they are likely clueful in other ways, too.

I'd say that the URL for Apple iPod is:

  http://google.com/search?q=apple%20ipod

and for Sony Walkman:

  http://google.com/search?q=sony20walkman

Really, I see people going to their GMail by typing "gmail" in Google. And ask 100 strangers on the street how important the tidyness of the text in their browser's address bar is to them.

Yup... and unfortunately for Sony, the top hit for "sony walkman" is Wikipedia. If their URLs didn't completely suck maybe they would have better luck with their SEO.

I think it's Wikipedia because "walkman" has a much longer history as a brand name, from cassette players to cell phones.

I remember times where back where I lived "walkman" was a plain noun meaning "any kind of portable cassette player".

Well, in a way the same thing is true with "iPod" now: It's used to refer to "any kind of MP3 player". But search for "ipod" on Google, and the Wikipedia page telling you all about the history of the iPod is not the first result, it's a page where you can actually buy one. Several pages, actually.

Hmmm. I don't believe people use the term 'iPod' to refer to any old MP3 player.

Some do (my parents for example), and I suspect even among those who recognise different brands, there's decreasing awareness of what MP3 is. It's not uncommon for a brand to become synonomous over time - Walkman was an earlier example in this thread, but others include Liquid Paper, Roller Blades, and even Biros. Dangerous territory, because in the instance where it does become generic your trademark protection disappears (see also the battle Xerox continues to wage in that regard).

Also, Kleenex. Possibly the "Duck" brand of tape as well since people seem to have trouble saying duct tape.

Actually duct tape vs duck tape isn't all that clear.

http://en.wikipedia.org/wiki/Duct_tape#Etymology

Duck Tape is trademarked, but it's possible that was the original term for it. It's weird.

Many of my friends do; also many older more "technically-challenged" people I know do.

I don't appreciate the downvote.

I know of no-one who does, technically challenged or otherwise. Also, I don't have downvoting rights.

I suspect Wikipedia's massive page rank helps with that. Pretty soon wikipedia will own most keywords.

You're right. URLs like Sony's are exactly the reason why people completely ignore URLs, and the only URL that any average joe ever types in the address bar is www.google.com.

Actually, no, it goes deeper than that.

Most people never learn how to use the address bar. Period. It's not that the URLs are too hard or complex to learn, it's that they don't know there's anything they could learn to type in there.

Well, part of the reason for that is that people never actually need to learn about URLs, Google is perfectly adequate for the average web browsing needs. And I would say most people know how to copy/paste URLs to share them, so it's not a problem of 'learning' what they are and what they're for.

Despite the google argument, URLs still count for sharing links with people. If you have to email someone a nasty unreadable 140-char URL, it depresses the shareability of that link in every way - digitally, orally and written.

Sony's mixing the store with the product information page, resulting in such an ugly URL. Regarding stores, however, Apple isn't entirely innocent either—what about URLs like this[1]?

Sony UK's is just a product information page, and it has a much cleaner URL[2] (albeit not as clean as Apple's).

[1] http://store.apple.com/us/browse/home/shop_ipod/family/ipod_...

[2] http://www.sony.co.uk/hub/walkman

I get the point he's trying to make, but you can still type in http://sony.com/walkman and it takes you to the correct page.

How do you discover that though? If you get to it from Google or another site, you're immediately thrown onto a page with a hideous, unrecognizable URL.

It's ironic that a post about URL tidiness would be here: http://reillybrennan.com/post/7686471401/even-a-url-tells-a-...

That's tumblr's URL scheme, and I don't believe there's any way for the user to change it.

Which might be a reason to avoid using something you can't control.

Aside from maybe removing the numeric ID, is that url that bad?

A better URL would be http://reillybrennan.com/2011/even-a-url-tells-a-story - stick the year in there to get a fresh namespace every 12 months so as not to end up with collisions some day.

I can see going either way with or without the year, but not for the reason you stated. If you are writing the exact same title for a blog post then the year should probably be in the actual title because that most likely is an important piece of the article (thinking "Resolutions - 2011"). The reason I might include a year in the path is just to let users know quickly how old the material is when they see a link to your article, but then again do you really want to let them know that before click into your site?

I like using years for namespaces because I like thinking long-term - in 50 years time, will that site's namespace be getting too busy? Using years means you can even completely change the rest of your URL scheme without needing to update last year's URLs.

Quickly letting users use the age of the article is a good reason too, but only for blog-style content. I don't think you should use the year trick with evergreen content (example.com/2010/terms-of-service for example).

example.com/2010/terms-of-service would still be a good URL for the namespace reforming reasons you cited.

Why include /post/ if the blog is on the root domain?

I like the point of this post, but I wish the author would have elaborated more. Sanitary URLs are great because they play on human emotions. Looking at the simple, clean Apple URL, it does two things: it provides an easy to remember route to their specific product, and two, it feels clean. In contrast, the Sony URL feels dirty and disorganized, lacking a real focus and seemingly, makes the Walkmen feel like just another electronic whatchamacalit. Although it seems purely cosmetic, there's a subconscious thing going on. Yes it helps developers/consumers but even more importantly, it makes your brand look focused. There should be more writing on the psych aspects of web development. I feel like there's a lot to be said.

Donald Norman wrote a great deal about that subject (psychological and emotional aspects of user experience), albeit not specifically about web dev.

I thought the name looked familiar. I downloaded a copy of "The Design of Everyday Things" not too long ago. Guess I'll have to crack it open?

For those that are also interested: http://www.amazon.com/Design-Everyday-Things-Donald-Norman/d...

And there's also Emotional Design: http://www.amazon.com/Emotional-Design-Love-Everyday-Things/...

The interesting thing though is that Jobs was always very inspired by Sony and "tidyness".

I really do wonder if Apple can continue without Jobs or whether it will decay into management kingdoms like Sony and other large organizations.

The story here is, 'hide complexity'? Or perhaps 'Focus on the details of user experience'?

I love the reference to "all of this business at the top". Perhaps there's a design principle in there. If a user is ever referring to "all this business..." then you've gone off track.

Who cares about urls? My browser hides everything except the domain anyway

Who cares about urls?

Clean URLs can help developers to better manage their site. This is because they provide an intuitive way to organize site content. By design, this also means that users can find information easily.

As a simple example, imagine surfing an exotic car website. You want to find information about the Koenigsegg Agera. Now, imagine the following two URLs which both point to the page about the Koenigsegg Agera:

1. http://www.example.com/vehicle.aspx?id=0x5F3759DF

2. http://www.example.com/vehicle/koenigsegg_agera

Clearly, the URL in example 2 offers:

1. An intuitive way to navigate the site's content. As a user, you could enter a different vehicle name, using the given URL structure, to locate other vehicles:

http://www.example.com/vehicle/ariel_atom

http://www.example.com/vehicle/porsche_918

2. SEO (Search Engine Optimization). This is a topic unto itself. In general, one of the areas of SEO is optimizing URLs in order to achieve the best results with regard to indexing and ranking. Please see Google's Search Engine Optimization Starter Guide (see link, below) for a more in-depth treatment of this topic.

References:

Clean URL http://en.wikipedia.org/wiki/Clean_URL

Google's Search Engine Optimization Starter Guide (See page 8, "Improve the structure of your URLs") http://www.google.com/webmasters/docs/search-engine-optimiza...

URL structure http://www.google.com/support/webmasters/bin/answer.py?answe...

Search Engine Optimization http://en.wikipedia.org/wiki/Search_engine_optimization

I went through your links to understand the SEO issue, and I am having a hard time finding more than one technical reason to make URLs friendlier. The only thing that stood out to me was along the lines of including searchable keywords in your URL, so indeed, rather than just an ID, having the word "walkman" in Sony's URL would raise their URL score in the search engine.

The rest was pure speculation on what "users" do with URLs -- aka some engineer's personal opinion on what's pretty. Can anybody provide links to usability studies regarding what users actually do with URLs? This would be of far more value than some company's personal standard, be it Google or someone else.

My experience is unfortunately anecdotal, but all my users just bookmark or copy-paste whatever they see. They are more concerned that the site has good navigation than that they can alter the URL or type it in directly. They use a starting point and then move on from there. So I have a hard time believing statements in these documents like "users may remove a part of the URL, thinking it is not important". The one exception is that I have seen newer users type "www." before URLs that do not need it.

(I am attempting to repond to both Xurinos and romaniv in a single reply here.)

My clean URL example was just an example. I did not intend to infer that the world is primarily navigating via URL manipulation, but it came across that way. Rather, I meant that it can be a nice touch for usability, in certain cases. For example, power users are going to quickly notice, "Hey, I can find a vehicle by manipulating the URL by using the 'manufacturer_carmodel' pattern." Of course, the site would have normal search functionality, so that traditional users could locate vehicles in that manner as well. In the end, it is more about SEO, but with the clean URL paradigm, you get both (again, in some cases), so it can be a win-win. Also, the URLs look nice when they appear in search results (i.e., nice keyword-based URLs instead of URLs with query strings and IDs) which can lend itself to readability/usability.

Of course, even with clean URLs, a case where it probably does not work is a blog (as noted by romaniv). For example, no one is going to manipulate a long URL string for a blog posting; they will simply navigate the site for the information. Although, here is a real-world example that I have encountered: I have come across multi-part blog postings via Google searches. In many cases, "Part X" displays in the search results, but not "Part 1." In these cases, it is always nice when these sites are designed with clean URLs, making it a breeze to simply copy the URL from the search results, and change the "X" to "1". Clearly, this is not easy (or possible) to do if the URL is a query string with an unintuitive article ID.

Ultimately, I think that clean URLs are about SEO, but I think they can have a usability factor, again, in some cases and well, for "power users" at least. I probably should have said "power users" instead of "users" in my original post.

An intuitive way to navigate the site's content.

This assumes the user knows the correct product name and the rules you use to normalize it (ariel-atom vs arie+atom vs ariel_atom). This not the case for most URLs. I don't remember the exact title for each article in some blog. In fact, I'm more likely to remember http://example.com/?Article.view.1 than http://example.com/articles/2010/01/02/This_is_my_article_s_...

Please note the underscore between 'e' and 's'. Websites usually normalize their titles when putting them into URLs, and they don't just URL-encode them, because URL encoding looks bad. They use arbitrary normalization and shortening rules.

Also, why would the user navigate via URLs instead of using website's UI?

I think URLs should be reasonably short and have no junk, but I also think that the whole "put-words-into-urls" movement is driven mostly by SEO considerations that have nothing to do with user experience.

>Who cares about urls?

Google does. If you have a non-descriptive URL, you're missing out on an opportunity for search-engine optimization.

Users don't, unless they are typing them. Which is rare, since most people use search engines and click the links. However if you are developing a URL that you intend people to type, of course you make it easy. A deep URL for a product in a shopping site? Users don't care what that looks like.

SEO is another matter, self-descriptive URLs are generally ranked higher.

I wrote a thing about my exceptionally short urls a little while ago: http://qntm.org/urls

IMO, this shows how important each respective product is to each company.

Reply on news.ycombinator.com