Computing desk
< May 12	<< Apr \| May \| Jun >>	May 14 >

Welcome to the Wikipedia Computing Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

May 13

mixing drive types in a raid

I'm thinking of getting one of those 4-bay usb/esata RAID enclosures and would be using drives of similar capacity but several different brands. The enclosure comes with a Seagate drive and I have some existing Samsung and Hitachi drives that I want to put into it as well. Is there likely to be a problem? Also, is it ok to use so-called "Eco" drives in a raid for personal use? I don't care if that makes it operate a little slower since it's mostly for archival storage on USB, but I wonder if it could impact data integrity. I do know the usual reason for "raid" drives is they time out faster in case of media errors, since they don't want to degrade server performance and they expect the raid itself to take care of error recovery. This would be RAID 5 (unless someone thinks I should really burn a drive worth of space and use 1+0) if that matters. Thanks. 66.127.55.46 (talk) 05:51, 13 May 2012 (UTC)[reply]

You can use pretty much any drive in a RAID array. You're running it over USB, so I doubt the bottleneck will be in the drives you select. I've run an array of mixed brand consumer disks in the past with no problem. The only thing I can see being an issue is that the different disks may not be exactly the same size. My drives varied by hundreds of megabytes among brands even though they all had the same GB number printed on them. My controller had no problem handling it - it just didn't use the excess space on the larger drives. With 4 drives RAID 5 should be fine. More redundant systems are useful when you're running enough disks that it is likely a second drive will fail before you've had a chance to replace the first bad one.209.131.76.183 (talk) 13:09, 15 May 2012 (UTC)[reply]

Is QR code really free for all to use?

The fact that the QR code technology is legally open-to-use is obvious, since it's patent has been expired (and Denso Wave has ignored the patent rights if it was still valid). However, ISO distributes the specification for a great fee. More than that, the spec is fully copyrighted and cannot be distributed without permission from the ISO (which you do not get any license in most cases). That means you must buy a "license" to use the work from ISO and it is technically impossible to implement QR without payment. Or you will have to use a library (you will have to comply to the license of the library, most are copyrighted). Or even traveling to 70 years after ISO's death - when the technology is out-of-date. What do you think? 123.24.100.217 (talk) 10:19, 13 May 2012 (UTC)[reply]

QR is "free" in that you can build and distribute applications that use it without paying anybody. The ISO standard is not free. In the same sense, I can make and distribute copies of the 1911 Encyclopædia Britannica, but that does not give me the right to freely use English textbooks to learn the language. You can use other sources for than ISO for QR, just as I can choose to learn English from other textbooks or people. --Stephan Schulz (talk) 11:04, 13 May 2012 (UTC)[reply]

There are two things to distinguish between here. One is the technique for making/using/reading QR codes. The other is the expression of the standard as a document sold by ISO for $225USD or so.

The technique of making the QR codes is not synonymous with the expression of the standard. Techniques are covered exclusively by patents. Exact expressions are covered by copyrights. (There is mucho fuzziness in these definitions, but let's ignore them for now.)

If the patent is expired, there is nothing to prevent one from re-expressing the technique of how to read/make these through either the use of the patent or the use of the ISO standard. You just can't distribute the exact ISO standard expression itself. Re-writing it in your own language, however, is completely cool by the standards of copyright.

Is it true that it is technically impossible to implement QR coding without the ISO standard in front of you? I doubt it very, very much. --Mr.98 (talk) 13:41, 13 May 2012 (UTC)[reply]

Here are links to 10 archived discussions which are more or less related to this discussion.

—Wavelength (talk) 14:46, 13 May 2012 (UTC)[reply]

Sometimes I can't tell if you're a bot or just confused about what the OP is asking about. Not one of those links come anywhere close to answering the OP's question. This is not a helpful response. --Mr.98 (talk) 14:51, 13 May 2012 (UTC)[reply]

Yes it's certainly possible to implement QR libraries without reading the standard, since there are many existing implementations you can refer to, among other things, so you can reverse-engineer the spec. You might have to read the standard if you were implementing some kind of certification suite for standard compliance. In practice most users will just care that your library works. It's also of course possible to read the standard without purchasing a copy, for example if your library has it. 66.127.55.46 (talk) 15:13, 13 May 2012 (UTC)[reply]

Sounds like we need a project WikiStandards where people could write up a standard that is more standard than the ISO standard, which is the first and final resource for all developers. Wnt (talk) 16:11, 13 May 2012 (UTC)[reply]

There are Open standards. I don't see the advantage of using Wiki software; the whole point of standards is that you freeze them from editing after they are finalized. Anyway, the amount of money we are talking about here is not very much as far as I can tell (a couple hundred dollars is peanuts in the scale of things). I don't think access to ISO standards is any real barrier for development. Patents are much more problematic. --Mr.98 (talk) 20:17, 13 May 2012 (UTC)[reply]

Getting more in to the patent issues, Data Matrix mentions some patents covering Data Matrix which I guess which would apply to QR code which are expired, I presume these are what the OP was referring to when they mention 'patents expired'. I would note I don't see any suggestion in QR Code that Denso Wave's patents have expired, simply that they've chosen not to exercise them which is supported by [1]. I didn't look in to the patent filing dates of grant dates but the history shown in the page of the first standardisation being in 1997 suggests it could be a few years before they've expired everywhere. However the terms of this 'not exercised' is unclear/unspecified, there may be more information somewhere, if it matters to you that what you're implementing is open and free, you'll likely want to look in to these. There is a reason why there's controversy in cases like Mono (software), Office Open XML or in general Microsoft Open Specification Promise and the case of Java [2] [3] about the terms of the patent grant and while complicated licences or grants like in those cases can complicate matters (e.g. only applying for implementations which comply with the standard), it can also provide assurances (e.g. making it clear it's for the lifetime of the patent i.e. irrevocable). The fact that QR Code is an ISO standard and that I presume Denso Wave may have made some guarantees there as well, at the very least to licence their patents on fair, reasonable, and non-discriminatory terms. Of course as history has shown you can never be sure when some patent troll or whatever is going to come out of the woodwork so even if you have assurances from Denso Wave, you can't really be sure you want have problems. As Mr.98 has said, patents are generally the biggest problems. Nil Einne (talk) 02:32, 14 May 2012 (UTC)[reply]

I think there is some confusion about what standards are for. They are, in general, almost useless for developers, except in the last steps of checking an implementation for conformance to the standard. They basically represent agreements between developers/vendors of already-existing implementations, who want their products to interoperate with each other, so they write down what various features are supposed to do. As such, their essential characteristic is sign-off by the vendors. In some cases the committees rather ruthlessly strip out any info about how to implement anything: they just say what what the output of something should be, not how to make the output (which can be difficult). E.g. the standard for something like mpeg video would give enough info to write a decoder (since it says what video output should result from mpeg input) but says absolutely nothing about how to write an encoder (the complete specification of an encoder is basically that it produces something that the decoder can play back, so the standard only describes the decoder). Also, some standards (not QR) are secret (only paid-up industry members can see them under confidentiality agreements): DRM schemes are a typical example.

In the ANS standardization process (at least sometimes), the standardization committee produces a series of drafts that members vote on until they reach a version that they submit to ANS as the official standard. That means the last draft and the actual standard in practice say the same thing. In some cases, the drafts are allowed to circulate online so people can see them for free (I think the ANSI C draft standard is online like that). The main reason to pay for an official copy is so that you could certify compliance, but it's quite possible to develop an uncertified but working implementation based on drafts.

It might be worthwhile to write some on-wiki specifications for stuff like QR, but "Wikistandards" is not needed for that--just write a Wikibook about QR codes. Wikibooks is a good repository for that type of info, and it's actually a better editing environment than Wikipedia in all sorts of other ways too. 66.127.55.46 (talk) 20:49, 13 May 2012 (UTC)[reply]

Fair enough; but there's a difference between a standard document, and a reference implementation. They fulfill different roles; and there are all sorts of reasons why a standards committee might choose not to provide a reference implementation (a few legal-ese issues spring to mind). For example, there is the JPEG standard, and there is the unofficial cjpeg/djpeg reference implementation that just happens to be unofficially written by the same group of people who sit on the standards committee. Officially, they are the "independent JPEG group" - "an informal group that writes and distributes a widely used free library for JPEG image compression." As their disclaimer officially states, "IJG is not affiliated with the ISO committee." It just so happens that most of the IJG people are also conveniently participant to the ISO and ITU committees.

For QR codes, there is at least one standard: ISO 18004:2006. As far as a reference implementation, at least one implementation is available under the LGPL: QREncode, with documentation. Nimur (talk) 18:22, 14 May 2012 (UTC)[reply]

cyber crime involving January Makamba

please let me know if the above mention person who has been appointed as a ministry of Tanzania goverment is wanted by home land security of USA — Preceding unsigned comment added by Mikemike2012 (talk • contribs) 10:51, 13 May 2012 (UTC)[reply]

There is no hint of anything like that online. Note that we have an article January Makamba, although I would not put any great reliance on it. Looie496 (talk) 18:33, 13 May 2012 (UTC)[reply]

Wow, our access to news from that part of the world is not good. I just barely found a source describing his appointment. [4] Google News indexes some articles about him though [5] none of which sound like he's on anybody's hit list. Wnt (talk) 20:16, 13 May 2012 (UTC)[reply]

Yeah, Sub-Saharan Africa is pretty poorly covered here and in the online news; I've lately been writing articles on Liberian national politicians because I have access to an archive of printed newspapers, and even a comparatively recent presidential candidate didn't have much of anything online. Nyttend (talk) 03:40, 17 May 2012 (UTC)[reply]

internet 300 yards away

Okay so i gots me a friend who lives on a ranch. His only interweb options includes the following:

satellite

Satellite is too expensive/limiting/slow to be a real option for his internet needs. He has on his property another house roughly 300 yards away which does high speed cable internet. The house is not in line of sight due to trees. I want to get internet from that house to the other house someway. I have already considered running a line for Ethernet but i think that requires a power line to follow it for repeaters - which is costly.

I had an idea that i wanted to bounce off y'll - drop 1 line from the house with internet to the house without internet (300 yards away) and then directly connect a wifi router to one end and a wifi receiver (like a USB card or something) to the other end. I have no idea if this can be done while maintaining a stable internet connection that is low latency that can be used for online gaming like World of Warcraft. Any other ideas out there would be great - thanks 98.238.132.145 (talk) 22:02, 13 May 2012 (UTC)[reply]

This is a long-shot, DIY sort of thing, but I wonder if this isn't actually an appropriate place for a cantenna? --Mr.98 (talk) 22:06, 13 May 2012 (UTC)[reply]

Dont think a cantenna will work due to the wall of trees filling the space between the houses. 98.238.132.145 (talk) 22:51, 13 May 2012 (UTC)[reply]

Yeah, you're probably right. Googling around suggests heavy trees are pretty impervious to WiFi, probably due to their water content? Either way... --Mr.98 (talk) 23:45, 13 May 2012 (UTC)[reply]

…or run a wire from one antenna to another around the tree… one takes decades to grow and one is manufactured in a matter of minutes (or even dig a ditch and bury a cable, the exercise will do you good, and the physical cable will improve the connection no end). ¦ Reisio (talk) 12:19, 14 May 2012 (UTC)[reply]

Something like Free-space optical communication on a tall mast? or cut one of the trees down? Vespine (talk) 23:51, 13 May 2012 (UTC)[reply]

How does the other house have high speed internet? If it's cable/phone, can you extend the cable/phone drop from the house with internet to the house without internet? And then both houses could get high speed service. RudolfRed (talk) 01:08, 14 May 2012 (UTC)[reply]

The house 300 yards away has cable internet but your friend can't call the cable company and get the same thing? how odd. Anyway, yeah, some wifi hack sounds doable. 300 yards is pushing it for ethernet but there are some other wired schemes that allow longer reach. 66.127.55.46 (talk) 07:02, 14 May 2012 (UTC)[reply]

I would go for a wired system unless you can erect a tall mast to gain line-of-sight. Trees block WiFi-type microwaves. (I know that because this reply is being transmitted by microwave over trees!) Dbfirs 08:12, 14 May 2012 (UTC)[reply]

You must be strong to throw microwaves over trees! ¦ Reisio (talk) 12:20, 14 May 2012 (UTC)[reply]

Who mentioned ovens? I did have to raise the micro-dish a yard higher because twigs from the highest tree were getting in the line of sight. Dbfirs 07:10, 15 May 2012 (UTC)[reply]

Or RFC 1149, with enough of them and only 300 yards to cover, it could be about as good as 56k! ¦ Reisio (talk) 12:22, 14 May 2012 (UTC)[reply]

Ethernet extender mentions some options, like back-to-back SDSL modems over a copper pair. If you can find some old 10BASE5 nics and cable that's long enough, that would be another option Ssscienccce (talk) 13:55, 14 May 2012 (UTC)[reply]

(EC) Have to agree with others wired sounds like the best option. It won't be cheap but I don't think it'll be extremely expensive. You shouldn't need additional power cables, just use power over ethernet although since you're installing the cable anyway, it might be worth looking if it'll be more effective, we're obviously only considering low voltage power here. From a quick search, something like [6] may work (not sure how weatherproof they are though) and it looks like there are options for under US$400 (not including the ethernet cable cost or cost for you to install the cable). You may be able to do something cheaper, particularly if you can build your own sufficiently weatherproof enclosures and look around for cheaper more hackish options (remember a ethernet switch can function as a repeater). From your earlier comments, I presume installing a cable is doable, there isn't a river or road between the houses or something like that. You can look in to other ethernet options although I'm not sure if anything will be cheaper/easier. Nil Einne (talk) 14:09, 14 May 2012 (UTC)[reply]

For high-speed gaming, an optical fiber would be ideal (and not too expensive for the fiber alone), but the cost of the boxes at each end, and the expertise needed to terminate the fiber properly, makes this suggestion a non-starter unless you know someone in the trade. Dbfirs 07:10, 15 May 2012 (UTC)[reply]

If there is cable 300 yards away and doesn't mind paying for two internet connections, he can have the cable company provide him a hookup. If he runs the cable properly himself, they shouldn't charge much for tying it into the system - just check with the cable company first to see how they want it to look at each end. They might require him to leave the trench open so they can inspect things before he buries it. He might also be able to get away with just running it from where the cable enters the first house to where he wants it on his, then using a second cable modem - I would check with the cable company to see if they'll let him run two modems on the same account. — Preceding unsigned comment added by 209.131.76.183 (talk) 13:23, 15 May 2012 (UTC)[reply]