Another example of poor planning is that each time we spell out web addresses, we are actually hacking a design flaw. A good namespace should read like a hierarchical folder structure: com.google.www/finance/public-companies/yahoo/flickr instead of a hacked structure that goes in two different directions at once www.google.com/finance/public-companies/yahoo/flickr.

I wouldn't call english-only a design-flaw, a short-sided decision that calls for hacks. I call it the only possible method. Having keyboards/browsers that could handle every possible key combination in every language in the world would be a nightmare and would be extremely costly. Also do you include every single language no matter how few people use it? Just as air-traffic is english only, the internet needs a single language for the most part as does most global industries. The number of phone numbers available has been or will soon be a problem in many areas what with the increased number of cell phones and other devices. The number of IP addresses is supposed to be exhausted in the near future and so a new extended addressing system will be needed.

My personal favorite is the width of roads, and how the space shuttle parts were designed and moved... http://www.astrodigital.org/space/stshorse.html

I'm going to agree with Kevin; road width is entertaining; especially in the old European cities where the buildings are sometimes so close together that you can't get a vehicle through at all. This is also a likely reason that most of the vehicles they create are so compact.

There is a phrase (unfortunately I only know it in Dutch) which mentions the "disadvantage of the restraining/slowing head start", exactly what you are talking about.

You see it all around us - for instance, in major software systems (especially in the financial world) that are now very difficult to replace with better, more modern systems.

I guess this is why you run Beta versions of new products, before you unleash it on the full market.

One of the most famous 'design flaws' (although some might argue that it verges on urban legend) is the touchtone telephone keypad layout. I've always understood that it's the opposite of the calculator keypad layout because the tone-recognition technology when it first came out could not operate effectively at the speeds at which people (especially people who were proficient with calculators) could dial the numbers.

A future hack: The clock and the calendar. Both are tied to this planet, and as long as we're tied here they'll work well enough. But the day we move beyond Earth, both will be very inconvenient because different planets have different rates of rotation and different year lengths. Moving even further out, we'll run into the problem of trying to figure out which frame of reference is the "most valid" one: Local time on Earth, local time in a distant solar system, or local time on a ship moving at speed between them. A precursor to the problem is the original definition of "noon" as "when the sun reaches its highest elevation in the sky." That worked fine when we lived in small villages and didn't interact much with distant ones, but as soon as speedy travel was invented it all fell apart. We had to hack the clock and add time zones to the mix, and now I have to add three whenever I need to figure out what time it is in New York. Synchronizing our watches is easy on a local scale, very hard on a larger one, and darned near impossible on the cosmic scale. I don't know if that's so much a design flaw as an unavoidable problem, but it's something that can't be fixed without constant hacking.

America was built for cars and Europe for motorbikes. As the Internet becomes more of a popular means to do business then there wont be such a demand for roads cause we'll all stay inside and stay all comfy. Adam

IPv4

the size of the seats and the aisle on american commercial airplanes

this post is offensive and lacking research.

The Imperial Mesurement System has been re invented into the SI System.Which in my opinion makes alot more sense for measurement and temperature. Living in the UK, we have one of the worst examples of this kind of thing, SI units of measurement for most objects, yet imperial for driving distances and road speeds, crazy.

The QWERTY system was never designed to slow down typing. It was designed to prevent the typist from presses keys that were next to each other, preventing jams. It had nothing to do with speed. People that have learned the Dvorak system have generally reported ~10-20% increase in speed, which becomes a negligible benefit if you consider the handicap imposed by every other keyboard they will have to use (presuming they are unable to change it to Dvorak).

Very interesting viewpoint. I agree that it's a great thing to plan for the future, but that can only be taken so far. Some of the greatest innovations in our society were hacked together. Just look at Linus Torvalds' first posting regarding Linux to the newsgroups. http://groups.google.com/group/comp.os.minix/msg/b813d52cbc5a044b?dmode=source There's something to be said for practicality. Ideally, the first kernel would have had support for every kind of processor, but he ran with what he had, and allowed the rest of the pieces to fall into place later. Maybe that's not exactly what your post is describing, but quick and dirty fixes have their place in programming. Sure, they might not be ideal, but neither are the situations in which they are necessary.

Have you tried learning to play piano? Pretty difficult, if you ask me. Why doesn't some musical keyboard manufacturer launch at least one von Janko keyboard? Do we have to resort to hardware hacks on top of a regular keyboard just to have fun playing music? See here for explanations and hacks: http://www.red-bean.com/~noel/uniform-keyboard/ http://improvise.free.fr/altinst.htm http://improvise.free.fr/images/bbbwsma.gif http://improvise.free.fr/beanbut/bean.htm

I agree with JS. IPv4.

"Dvorak is faster" is an urban myth. It is not faster, it was not designed to be faster. I type Dvorak for 3 years now, it has nothing to do with speed. Dvorak is more comfortable to write, because I can say much more on the home row and I alternate much more between left and right. Dvorak is more ergonomically than QWERTY, not faster. It was never about speed. The second urban myth is that the QWERTY layout is for slowing people down or to prevent that keys get stuck. This is written on thousand pages on the net and it's uterly wrong and not backed up by any historical document. In fact, the person, who is most likely responsible for the success of this layout, because he firstly adopted it on the first really commercial successful typewrites (there were typewrites before, but they had all kind of different layouts) once said in an interview a long time ago (he's dead for ages, of course) "The layout is more or less randomly chosen". This is the best historic explaination we have for the layout, which is actually backed up by still existing documents.

Agrees with my blog posting <a href="http://blogs.cnet.com/8301-13507_1-9732586-18.html">Which keyboard</a> about QWERTY vs. DVORAK.

Time waits for no man to sit back and idolize. Without hacking and prototypes, you would not have seen the tremendous progress we have today. Sure, it's not perfect, but nothing is. Read a book on refactoring and agile development.

I don't think you can have this discussion withotu mentioning Path Dependancy. http://en.wikipedia.org/wiki/Path_dependence

This article is another example of "Can't see the forest for the trees" blindness. The premise is that success would be so much better if every product were designed with unlimited scalability. The REALITY is that success is so elusive that designing for limited scalability makes much more sense on a day-to-day basis!

If you're doing a thousand projects and ONE of those thousand will be a huge success, why should you expend 10-30% higher resources on all thousand projects when only ONE of them will warrant it? That means that each project will take longer and cost more which only delays your eventual success even more.

See, it isn't the projects that fail that bring you happiness. It's only that ONE success. So spending extra time and resources on failures or marginal successes "planning for the big time!" is completely counterproductive and might push you to just give up altogether.

It is actually better to design for ruthlessly fast prototyping and concept development because "getting to the big time" is much more important than having an easy road once you finally have the problems associated with being a winner.

Where are the studies that indicate the Dvorak layout allows a greater, mean typing speed? Very few (if any) technologies are perfect on inception. "Hacking" as you describe it is just plain "engineering"....it's WORK. It's quite easy to sit back and say in hindsight - "It should have been this way!" I am not a native English-speaker, but I recognize that there's an edge and natural predominance to anything that comes first. Maybe in decades to come, English will not be the prime language of the internet. Incremental change is effective and smart unless you've got something totally new and much much smarter.

If someone had a time machine and went back to create the perfect HTML standard, it would not have been so successful. The rewritten HTML standard would not be as forgiving and probably a lot more abstract. Most of the success behind HTML was that it was possible for anyone to create a (bad) web site, and still expect the browsers to display it.

But a far more important question is how we are controlled by convention. Take this blog for example. Why does every blog ask for your mail address? What use is it to you? I do not like filling in my mailadress, and I always use a bogus mail address.

You wrote: "The Dvorak keyboard ... will never be adopted by the masses". Maybe you are wrong. The generalization of keyboards with configurable layouts such as the Optimus Maximus (http://www.engadget.com/2007/11/28/optimus-maximus-in-minimus-sub-1000-configurations/) will certainly facilitate the adoption of the DVORAK keyboard.

The example of non english URLs is something only related to IE, both Safari and Firefox on Mac seem to use the Japanese 8bit characters in the URL. IE uses the URL encoded form which is unreadable. This is not a flaw in the design, because it works really well ... if implemented. The actual design flaw lies within the DNS system as this does not accept 8bit characters. Hopefully they will have IDN (International domain names) available and these will accept 8bit characters.

Here you can find examples of Japanese and Chines URLs in a browser: http://farm1.static.flickr.com/231/493610815_8e77241b56_m.jpg http://farm1.static.flickr.com/231/493610815_8e77241b56_m.jpg

K- interesting point. We actually ask for your email address so as to be able to contact you off list if you ask a question which bears a personal reply instead of a public one. I figure most people will use bogus addresses and I don't mind at all.

New Orleans - do you know how much it would cost to import enough soil to raise an entire city 20 feet? Your suggest a design solution that was simply not feasible at the time of construction, even if they had foresight. So this example is a non-sequitur. English on the internet - the limitation of domain name registrations to western characters is imposed for security reasons. There is really no technological reason it could not be expanded, and the regulating body has considered doing so many times. But other languages use characters that are similar in form to many western characters. If you allow for use of those characters in the registration of domain names, someone could feasibly register ebay.com replacing one letter with a similar structured foreign character, and spoof the site for data mining reasons. So it seems that the restriction of valid characters is not a shortcoming, but a security benefit. QWERTY Keyboards - I think others have shown this example to be flawed. The principle is valid, but this post does not support that principle with one valid example.

"Why would original developers choose to build a city on lower ground than sea-level in an area prone to hurricanes? Why not import more land to raise the city above sea level?" The land has been sinking for years. New Orleans didn't start out on land below sea level. And the city wasn't planned. It was settled.

hey, you forgot the uttermost design flaw... the decimal numbers! it's really sad we just can't reinvent the numbers, because 10 is just the wrong base for math, time and binaries. this is a great page on the subject http://www.infoverse.org/octomatics/octomatics.htm nagash

Um, I don't know what IPv4 is, but it sounds like a good example to me. I was just going to say Windows in general - basic design flaw

To those of you who worry that we will run out of ip addresses, fear not. We may have more people and more devices than ever on the net, but more and more of us are behind NAT routers anyway, so, for instance, my 3 currently online computers show only 1 public ip address to the net. The devices have local addresses assigned by my NAT (network address translation) router. Many businesses have hundreds of computers behind only 1 public ip address. We are in no danger of running out of addresses. Huge blocks of them are still reserved and not in use.

You hit the nail on the head speaking about New Orleans, that description blew my mind damn good article

Is New Orleans truly an example? I am wondering because it seems I read somewhere that the city has been sinking over the past 200 years and in fact was not below sea level when settled/built. Maybe I'm wrong...but I'm curious if anyone knows.

The metric system is superior to USA measurements.

Never realzied there was such poor planning going into major things. It really sucks :(

I disagree with the basis for this blog entry.

Lots of things reach mass adoption. When they don't, however, they deserve to, for whatever reasons inherent with the "properly redesigned" New Thing.

The core of the issue is a simple two words - WHO CARES. I don't mean that insultingly, I mean it as a simple question: How many people care about "poor design" enough to switch from an Old Thing to a New Thing?

Think about it -

Why hasn't Dvorak keyboard gotten anywhere? Because there's little need for a new keyboard layout, and 95% of people either type well home-row style, or they hunt-and-peck and it works fine. The other 5% had an issue, but {{who cares}}?

English on the internet? It's not a design flaw, it's how langauge is spread. English is common on the internet because enough of the world's Important Things use it. If someone is annoyed by the lack of their language, and don't want to learn English... {{who cares}}?

Mass Adoption is the greatest litmus test there is. If it's good, it becomes adopted by the masses. If it isn't good, us masses won't use it, and it'll go there it should go - to the Bin Of Obscurity.

Path Dependance :) http://eh.net/encyclopedia/article/puffert.path.dependence

Complete hack: The imperial system of measurement

Thinking about it, isn't more or less everything Humans have every achived a hack or progression? I can't think of a single thing that's ever been created that's been so perfect from it's initial creation that no-one has ever needed to change it afterwards. Except of course, the word "fu*k". That's pretty perfect.

Very confused about your Wikipedia example, as the address shows up perfectly with Japanese characters. If you're talking about the "wikipedia.org" then that could be seen as a different issue, as I think it comes down more to consistency. There could be a Wikipedia japanese domain - I don't mean just wikipedia.co.jp (not sure what jp org is) but then you'd have somewhat of a brand nightmare. Or am I just being an English weighted arse? ????????-.??? ?

Seems this blog doesn't accept Unicode input either :(

AC power plugs were earlier designed when more voltage was needed. Nowadays we don't need this much power to power most household objects. This is why most things have adapters to decrease the voltage. This heats up the adapter and wastes a lot of energy.

AC power plugs were earlier designed when more voltage was needed. Nowadays we don't need this much power to power most household objects. This is why most things have adapters to decrease the voltage. This heats up the adapter and wastes a lot of energy.

One word: Windows. Mass adoption and a lot of problems. They tried to redesign it, but every time people hate it and get pissed when they have to move to a new system. Old code, and lot's of it, but they can't get rid of it, cause Joe Schmoe wants to be able to run his powerpoint from 1849 cause he doesn't want to figure out how to use a new and much improved interface. Sorry, but has anyone noticed how much Vista looks like OSX? Expose? Cmon, they aren't even trying to hide that they are copying.

My examples would be:
1) Road System... I'm a motorbike rider and there are road signs and metal bariers to help people in cars survive on the road... both may kill a biker while he looses controll over the machine and falls down (and then slides for xx meters on the tarmac)
2) Windows 98 was a pretty good example what happens if you follow the "hack" path of development. You cannot build a skyscraper on top of your village house, neither can you build a reasonable GUI system around DOS.
3) Beta vs VHS system war shows that the quality of the system has little to do with its adoption by masses. People usually cannot see & understad the differences. The one with better PR & marketing wins
4) Thousands of electronic devices are capable of doing quite smart things but noone tries to implement the obvious software solutions. Do you have a cell phone? Does it have the onboard camera? Try to use it in skype via cable or BT on standard nokia / sonyericsson soft ;)))
5) Total lack of unification: left and right side driving. Can you see the scale of problems it generates nowadays? We have got to live with it, but it is both costy and very inconviniet. I do not think UK back in, say 1920 predicted the scale of problem. The same case 230V vs 120V, and finally the best case... In Europe we have around 15 different power plugs systems. Some are intechangable, most are not. Meanwhile European Union revealed the Directive on banana color & curvature allowed to be sold in EU and did nothing in terms of plugs. Moreover - nothing is being planned!