Self-Service Kiosks are not all that complex as you have explained here.
Maybe Self-Checkouts need some time to evolve wrt HCI but speaking of other suck devices such as ATMs they have nearly replaced the Teller completely for banking transactions.
But I agree making a machine to become 100% human-friendly is not as easy as it seems, i still find people struggling with their cell phones just to get used to the touchscreen, at times.
The best thing that governments do is screw up whatever they touch. Should the UN make a office software, you can bet your bottom dollar that they would have spyware embedded into the core. The best way to distribute the wealth is to make it yourself.
I suspect a large part of the problem with OO’s slow uptake is its price (or lack of it). Many customers equate price with value - otherwise how could Nike sell shoes at the prices they charge. If OO was sold as a commercial product at, say, 60% of the price than MS Office there would probably be more market up-take and there would be cash to pay programmers.
IMHO OpenOffice sucks. Sorry, but it’s extremely awkward to use; getting harder with every release methinks. It looks ugly, it is over-designed, etc.
I’m not saying that Microsoft Office is any better, but MS at least tries to make it better - one can argue how successful they are at doing it, but they try. I fail to see that OO even tries!
Whenever I try to make a presentation, a table sheet, or a large text document, I always start doing it in OO… and finally get enough frustrated to give up on it. Usually I end up doing it in a completely different program. Not MS Office (certainly not), but something completely different.
E.g. recently I did a lot of stuff in Apple iWork (Pages, Numbers, and Keynote). These products might still be quite limited compared to the competitors on the market and the performance is still a joke (don’t expect to make tiny changes to 50 pages document in Pages… it will freeze your whole system for 30 seconds!), but they have potential.
It always takes me a whole lot longer to go through the self checkout; mainly because the software and scanners etc. are all unreliable and poorly integrated. Those things are a design mess.
Also, there is a reason that human jobs are specialized-- practice makes perfect.
There’s lots of things in Open Office I’d like to fix, really trivial stuff like it doesn’t do section numbering properly. Word used to beat it hands down before the latest version where styles became a really confusing mess.
I like o-o for the only show the top-level style not every trivial bit of italics you’ve ever done, which is what Word used to do.
I hear that an o-o build can take a hell of a long time because it’s one monolithic thang, not properly modular. I’d love to make the word processor less nasty, but not if I have to screw around building all the other modules I never use and don’t care about.
If I could find something (not Abiword, which isn’t powerful enough) that just did, say, Word 95 with the multiple views of different parts of the same document that 2007 does I’d be content.
Jeff, perhaps you are missing the point of OSS. Take Asterisk for example. Created by Mark Spencer because he needed a PBX for his company. Buying the solution was too costly, so he coded it himself. Asterisk is a disruptive technology, so it makes sense it will garner an active community. Likewise, how fun can it be to code an office suite? Would you blame a fellow programmer for not taking their own free time to code something that might be boring?
I’ve noticed that some of the self-checkouts have been set to be more trusting taking into account things like I’m only buying one item, so I don’t need a bag and I can hold it while I’m inserting my money.
Speed is definatley another factor with these devices; I find that I can work A LOT faster then they can.
Having a self-checkout at stores like Home Depot is almost a joke. Let me left that 50lb bag of ReadyMix to scan it…
While forcing users to scan each item, one at a time, and force them to be weighted is a pain, here is my REAL pet-peeve…
A local market checked their lookup feature from
’type the first three chars’ to
’select a picture of your item from the list’ with 26+ pages of images in alphabetical order.
OMfG! what a major step backwards!?
I, the consumer, am able to type AND read!
Why throw me back to the stone ages?
THE major reason I check myself out is that I can do it FASTER than the cashiers!
-sorry to waste my shopping rant in a thread about open-source hurdles
The REAL issue with F/OSS, which a few comments have hinted at in passing, is what developers seem to forget about behind the F/OSS flag of Get the Source and Fix It Yourself Faster than Proprietary - you’re putting maintenance and bug fixes on your customers. While this might work for enterprises that can afford to hire developers to write and maintain this software, the premise falls flat on its face once you reach the consumer market.
Asking Aunt Mabel to code up a bug fix because she can’t print a picture of her grandkids that her daughter sent her is not just ignorant, it’s downright absurd. There is a reason people buy/download software - it places the responsibility of learning a programming language, debugging, etc. on someone else so they can just use it.
Aunt Mabel shouldn’t have to try to learn C++, find the API for her particular printer driver model, and code herself up a custom wrapper just to make OpenPhotoViewer 0.3 function at the level she should expect - click the Print button and the picture pops out in due time.
I’d love to use the scan-as-you-pick services many big stores have if it wasn’t for the fact that it requires registration and that they after this keep a list of EVERYTHING you’ve bought using this service.
Only to be able to send me more advertisement of stuff I’ll probably want to buy.
What I can’t figure out with the self checkout is how to use a freaking paper bag instead of a plastic one. If you put a paper bag on right away, it yells at you that a new item was placed in the bagging area and must be removed. And if you try and add it as you place the first item in, it tells you that the weight doesn’t match.
The only way I’ve found I can do it is by having the person working override it, which is a pain in the ass. So alas, I ussually use plastic.
You seem to be seeing more problems than there are:
What if the item you’re scanning isn’t found, or can’t be scanned?
I’ve never experienced an item not being found, so this must be very rare. If it can’t be scanned, I believe you can enter the code by hand.
How do you handle coupons?
I’m actually not sure if these can be scanned, because they need to be taken or voided manually.
Loyalty cards?
Scan it if it’s the keyfob type, or swipe it if it’s the credit card sized type.
Buying 20 of the same item?
Scan all 20. It might take a little longer than scanning one and entering the quantity, but given that the scanners these days generally work well and scan quickly, how much more time does it take than scanning one then still having to move the other 19 to your bags anyway?
How do you ring up items like fruit and vegetables which don’t have UPC codes, and have to be weighed?
Put them on the scales. Select the item from the categorised menu (with pictures).
What about unusual, awkwardly shaped items or oversize items?
Use the hand-held barcode scanner next to the screen.
He’s not having more problems than there are. He’s enumerating the error conditions that some user can possibly have. They are complications that someone who does not know the system is, at the very least, going to have to spend some time poking through their interface to figure out.
My problem with self checkout is when it’s misused. Like at Home Depot. I have almost never left home depot without at least one item that will not fit on their scale. And my local home depot has decided to close all of the service lanes and just pay four people to stand around looking at the self checkout. One of the worst examples of poor management I’ve seen.
Also, I’ll echo the statements about learning C++ or Java being a barrier. Anyone graduating with a CS degree knows one of those two - they’re the lingua franca of programming. Now if you make me learn Ruby, I’ll probably thank you afterward, but it’d be much more of a barrier initially.
@James Don’t be ridiculous, nobody expects an unskilled user to maintain their own software. There are lots of people working to make Free software as user-friendly as possible.
I hate the hassle of trying to buy beer at a self-checkout.
Jeff, another great analysis. Probably the most salient point yet about why self service checkouts are something I love even as others hate on them.
I work with POS software daily as a reseller/support person and you’re right on target. Thanks for a brilliant piece again.
@kris: Newer versions of the NCR software will beep at you when the transaction starts and ask you if you’re using your own bag, then re-zero the scale.
Josh (3rd post or so),
How in the world could an OSS project have bugs waiting for developers to fix in the amount of detail you list? Who’s going to find the source file containing the bug and attach it to the bug? It would take a developer to find the source file. At that point he may as well make the fix himself. 80%+ of all bug fixing is the research. The research is the painful part of bug fixes, the part that’s unsexy. Who in world wants to suffer all the pain and not carry through to the 20% of the problem which rewarding: fixing and checking in a bug fix?
And who is writing these specifications? Again, if you have someone around who can write meaningful specs, that person most likely can implement them too (as you noted).
Tutorials on how to contribute: Who has time to write these things?
I have always been perplexed by the elderly couple that routinely ops for the self-checkout, only to spend about 5 times as long as they struggle to understand the prompts.
They always look around as if to say why are they doing this to me?
Did they just not know it was the self-checkout line? Do they think all the lines are that way? I get so frustrated. What in the world are they thinking?
One problem I see with OpenOffice is that all they are doing is trying to ape Office. Office is cheap enough and good enough. Heck wordpad is good enough for most jobs. Why give time to that?
Do something new and fresh and better and they will come.