Wikipedia:Reference desk/Archives/Computing/2009 August 21

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.

August 21

When was the first Whitehouse email?

I'm trying to figure out when the first "email" was sent or received by the Whitehouse. I've found lots of other milestones (first email, SMTP history, whitehouse website history, etc.) but not the actual date that a president, or their whitehouse staff, had access to email in the building. Ronabop (talk) 06:13, 28 August 2009 (UTC)

Looking for a nutrition web site

I originally posted this Q on the Science Desk: Wikipedia:Reference_desk/Science#Looking_for_a_nutrition_web_site. It was suggested I cross-post to here.

I'm looking for a site which will list foods which are good for diabetics, those with high blood pressure, those with kidney failure, etc. No big deal ? Well, I want it to recommend foods for those with multiple medical conditions. (Note that I'm not asking anyone here to recommend any foods, just a site where I can look it up myself.) Ideally you should be able to weight the medical conditions, say if you only have slightly high blood pressure but have severe diabetes. In that case avoiding foods with a high glycemic index would be more important than avoiding salt. It should also list foods from best to worst, for the given medical conditions and weighting factors.

The Best Buy web site can do something similar with appliances right now. You give it some parameters, like the size of monitor, it's resolution, and cost range, and they give you a list of monitors that match, then they can sort by, say, the current price. The same thing could easily be done with foods, where the parameters become low glycemic index, low salt, etc. The site nutritiondata.com already has this info, I just need it to be made "searchable" and "sortable". So, does such a site exist, or must I create it myself ? If so, is there a way to extract the data from an existing site and add searchable and sortable columns to it ? StuRat (talk) 00:44, 21 August 2009 (UTC)

Categorization tool/bot

I want to add several categories to existing articles on a major scale. For instance, adding [[Category:Gram positive bacteria]] to all Staphylococcus articles. Is there a tool or bot that can help? I've searched WP:BOTS, WP:TOOLS, and the searchable archives for this talk page to no avail. - Draeco (talk) 07:22, 21 August 2009 (UTC)

WP:AWB. ---— Gadget850 (Ed) ^talk 12:50, 21 August 2009 (UTC)

Mediawiki error

So I have been getting this annoy error, it first popped up when I tried to use the Social Profile with Mediawiki 1.15.1, then in a desperate attempt to see if it was because the software was incompatible due to age or something I updated to 1.16 and then when that failed I did a shits and giggles attempt and installed 1.16WMF (which comes with the most recent version of the extension). The error is "Warning: Parameter 1 to incEditCount() expected to be a reference, value given in ****\w\includes\Hooks.php on line 132"

The extensions I used can be found here.

This error happens with ANY AND EVERY extensions that has to use Hooks.php. Please note that I have deleted the Hooks.php and replaced it a number of times and the error makes editing impossible by happening after an attempt to confirm an edit is made.Rgoodermote  07:28, 21 August 2009 (UTC)

Is there a search engine or program that can manipulate row/column arrays?

Many web sites have tables of numeric row coumn data. For example, imagine a nutrition table with rows of foods and columns of nutrient contents for each food. Does anyone know of a program that could manipulate any recognizable website data array like a spreadsheet, so that you could sort the rows by the values in a specific column? alteripse (talk) 11:50, 21 August 2009 (UTC)

Note that this is related to my nutrition Q, 3 questions up. StuRat (talk) 12:04, 21 August 2009 (UTC)

It sounds a bit fiddly or problematic without <td id="???"> or <td name="???"> tags - is it possible to assume that it uses tags such as these (ie the standard HTML Document object model)?83.100.250.79 (talk) 12:54, 21 August 2009 (UTC)

It would be trivial to write a greasemonkey script to do this with the tags, less so, but still possible without83.100.250.79 (talk) 12:56, 21 August 2009 (UTC)

If you are writing the web page from scratch then it should be easy to use javascript to create a sortable table using the tags described above.83.100.250.79 (talk) 15:48, 22 August 2009 (UTC)

Unfortunately, the idea is to extract the info from an existing web site, and we can't rely on them having those tags there, either. StuRat (talk) 19:39, 22 August 2009 (UTC)

MySQL -- from scratch to database in how long?

I know this is highly subjective, but I'm just looking for a rough estimate.

My personal background: master's degree in science, HTML literate, learned CSS this past Spring without difficulty, comfortable in Dreamweaver CS4, no real programming language experience, maintained a personal website as a hobby for years

My goal: Add a searchable, sortable, expandable post-launch database with upwards of 200+ binary (yes/no) variables for each entry to my existing website.

Tools at hand: Learning PHP, MySQL, and JavaScript: A Step-by-Step Guide to Creating Dynamic Websites by Robin Nixon + whatever info I can find online.

Basically, this dbase has been a dream project of mine for a few years and I've finally got time to pursue it. Hosting is a non-issue, as I've already got a lovely plan with all the extras I need. I'm here to ask about time. Not knowing a lick of MySQL, I am wholly unable to estimate how long it will take me to get up to speed and how simple/complex my dbase idea is compared to other MySQL applications (I'm guessing "not very")... the upside of the project is that daydreaming about it for several years has resulted in a rock-solid task list, with no possibility of feature creep.

Anyone out there wanna take a shot at this? 100 manhours? More?

Thank you, 61.189.63.183 (talk) 12:03, 21 August 2009 (UTC)

If you want an accurate estimate, you need to give us that rock-solid task list. For example, how does the data get entered, updated, and accessed ? All through the web site ? And how much data is there ? StuRat (talk) 12:09, 21 August 2009 (UTC)

Yes, more info needed. The hardest part about a database is getting the data in it. If, for example, you already have the tables filled out in Excel spreadsheets, turning that into a functional database takes not very long at all. If you have to put together complicated data entry tools, though, that takes 90% of the time. MySQL itself is generally dead easy, because you aren't doing that much with it (add a row, get a row, compare value X to value Y, etc.). It's the PHP, Javascript, etc., that you are making work with the MySQL that takes time. The database is the easy part—it's just a bunch of values in a table—the application is the time-consuming part. --98.217.14.211 (talk) 12:40, 21 August 2009 (UTC)

OP here. I was planning on having the data updated via a form on the website, but could certainly do it via Excel spreadsheets. This is the part of the plan that I can't speculate on because I don't have any experience with anything other than HTML & CSS. As far as data structure, each entry is a unique name and then a simple has/hasn't checkbox for a LOT of variables. I'm guesstimating 200+ right now, but each actual entry will probably never have more than ~40 checked boxes (yeses). Because the data entry will be brutal no matter how you slice it, I was hoping to make that available online for multiple people to input together. So I guess the database would have 2 access methods - browsing & editing/updating. As for searches, I just want visitors to be able to check as few or as many of the same variables as they wish, and then have the dbase spit out every entry that contains those variables. (chocolate AND vanilla AND nuts AND low-fat...) Because everything will be binary (yes/no), I'm imagining this aspect being relatively simple. Please ignore any urges to get sidetracked about the usability of the system - that can come later! Right now I'm just looking for a rough estimate. More info, if necessary, can be provided. 216.93.191.240 (talk) 13:14, 21 August 2009 (UTC)

In my experience, unless you have a bunch of slaves who are happy to do the data entry no matter how you set it up, you will be best served by spending some time thinking about what an intuitive way to enter all that data in would be. Hint: clicking on Yes/No checkboxes is NOT a good way to do it -- it is time consuming and much more difficult than, say, keyboard-based approaches. Coming up with a good interface is a MUCH harder programming and design task than the database.

Your database structure is pretty simple (though there are a few different ways you could do it—depending on the specifics of the data, which I can't tell from your description). Setting up a MySQL database can be done very quickly if you get a GUI program for it, like phpMyAdmin. You could probably have the basic data structure up and running in a few hours. Setting up the PHP to access the database is pretty straightforward if you follow the instructions in your book (and don't try to just guess your way through it—there are specific ways things are done and orders things should be done it). So I would say, setting up a bare-bones proof-of-concept probably would take, oh, maybe 20 hours total, assuming you are able to learn and debug your work at a modest pace. (For someone more familiar with the tools, it would take maybe 1-2 hours, but there's a learning curve in this as in everything). Developing something more user-friendly, that would take maybe 100 hours, assuming you are not a Javascript whiz already (Javascript is your friend for creating good interfaces, but it takes a little time to learn how to use it appropriately as well). --98.217.14.211 (talk) 16:42, 21 August 2009 (UTC)

Thank you for taking the time to post, everyone. 216.93.191.240 (talk) 23:02, 21 August 2009 (UTC)

Word Processing - half a line

In the usual suspects, is there a way to easily insert half a line feed. eg between two paragraphs to have a blank line that is only half the standard blank line. Specifically I'm thinking is there a way to make it easy to insert as well eg CTRL+Enter = Line feed 1.5 ? 83.100.250.79 (talk) 12:10, 21 August 2009 (UTC)

In MS Word, the way you'd generally do this is select the paragraph in question, then go to Format > Paragraph. There is a section that says "Spacing"; set the "After" to whatever the equivalent of 1.5 lines is at your font size (it is in points; so if it is 12pt, try 18pt). --98.217.14.211 (talk) 12:44, 21 August 2009 (UTC)

Thanks, I can't find an equivalent in open office - is setting "spacing below paragraph" to be >0cm the way to do it? Or have I missed a better option?83.100.250.79 (talk) 12:51, 21 August 2009 (UTC)

Yes, it is probably that. --98.217.14.211 (talk) 17:02, 21 August 2009 (UTC)

Directly below the "Below paragraph" is "Line Spacing" and a drop box where you can choose 1.5 lines. -- kainaw™ 14:32, 21 August 2009 (UTC)

I want extra lines after the paragraph, not in it.83.100.250.79 (talk) 14:35,

Then, increase the spacing either "Before paragraph" or "Below paragraph" as mentioned. You can alter the style itself so you don't have to manually do it to each paragraph. -- kainaw™ 21:07, 21 August 2009 (UTC)

21 August 2009 (UTC)

More:line feed without carriage return

A way to do this ?83.100.250.79 (talk) 12:51, 21 August 2009 (UTC)

Vertical proportional printing/justification

Is this possible (eg within a text box), does it have a proper name?83.100.250.79 (talk) 12:51, 21 August 2009 (UTC)

HDR

How do High dynamic range imaging and High dynamic range rendering work? The respective articles are a bit too heavy on the jargon for me. 69.77.202.90 (talk) 13:23, 21 August 2009 (UTC)

The short version is that HDR combines three images taken at different exposure settings. Using photos as an example, most digital cameras have a range of five stops - which means that you can have detail in the dark areas or detail in the light ones, but not both at once. Thus the camera normally picks something in the middle, allowing the mid range to look good, but not the two extremes. You normally see it with dark areas where you can't really make out what's in them and a "blown out" sky that just looks like an expanse of white. With HDR you take three pictures, one with a short exposure showing the light areas, one with a long exposure showing the detail in the dark areas, and one in between (the photo you would normally take). Then you combine the three, giving a picture that shows detail throughout. - Bilby (talk) 13:44, 21 August 2009 (UTC)

The rendering (of three photos into one) takes place by comparing the three photos pixel by pixel, this is why the three must be EXACTLY the same shot, at various exposures. The minimums in the short exposure (the too dark shot) and the maximums in the long exposure (the too bright shot) are thrown away and replaced with a more balanced value from one of the other two shots. The process is deceptively simple, but obtaining a good balance of what is 'too dark' or 'too bright' is sometimes difficult, leading to advantages in certain applications that use more complex algorithms. This is why HDR is most often used as a silly effect filter as opposed to a serious technique to improve a scene. --66.195.232.121 (talk) 13:54, 21 August 2009 (UTC)

(This is the answer for High dynamic range rendering, the first sort is different and given above.)

A typical computer monitor has ~256 scales of grey going from 0 black to 255 white.

In the real world light intensities vary much more eg the light in the sun is 1000 times brighter than indoors light (eg a lightbulb looks very dim when turned on during the day)

High dynamic range rendering calculates the light intensities at the higher scale (ie a big range of numbers typically with a intensity range of greater than 1 to 1billion actually even more)

However though the computer can stored these numbers (ie the calculated light intensities) for each pixel, a computer monitor can't display them.

To get round this the values are scaled back to fit inside the 0-255 range. (basically divide by a big number so that the most intense lights match 255)

(The scaling isn't always linear - can be logarthymic.) - in fact the way the big light intensity values are scaled is an important part of making computer HDR work well.

Ask if you require more expansion.83.100.250.79 (talk) 13:49, 21 August 2009 (UTC)

Windows mail password prototection?

My wife and I both use the same login profile for windows vista. When I (or she) click on 'Windows mail' my inbox automatically comes up. She logs into her hotmail email via the web. Is there any way I can put a password on windows mail so she can't get into my email. I love her dearly but I need a little privacy for the next few weeks. I don't want to give her a whole new vista profile to log into (that might make her suspicious) but I would like to stop here accidentally getting at my mail.58.170.109.8 (talk) 13:55, 21 August 2009 (UTC)

Sorry, but as far as I know you can't. Because Vista (and XP for that matter) are setup with the ability for people to make their own Windows user profile this is why the option doesn't exist in Windows Live Mail as it's expecting the security to be handled by the login itself (i.e. password protect the Windows account). ZX81 ^talk 14:16, 21 August 2009 (UTC)

Just remove the account info from Windows Mail (if possible) and use webmail for the next weeks? Or start a Gmail account to get all the info about the super-secret surprise anniversary vacation (or whatever your secrets are) sent there instead? Jørgen (talk) 17:45, 21 August 2009 (UTC)

Install Windows Live Mail. It's an upgrade for Windows Mail, and you can set it so that a a password is required everytime. Warrior4321 19:18, 21 August 2009 (UTC)

Another alternative might be to move windowsmail.exe (or whatever it's called) from wherever it lives now to C:\super\secret\location. Then clicking the icon won't work, but you can still go and manually start it. --Sean 19:32, 21 August 2009 (UTC)

The path is "C:\Program Files\Windows Mail\WinMail.exe". But moving the file might not work (you can try, though). However, my personal opinion is that every person using a computer should have her own user account. Always. --Andreas Rejbrand (talk) 20:38, 21 August 2009 (UTC)

I agree - it's not really even about privacy, it's about accuracy. If the computer says "Joe has mail", you won't know whether Joe or Joe's wife, daughter, cat, or sentient robot (who share the same account) actually need attention. The computer has no way to distinguish individual users, except in the form of login-accounts. If you want to make data shareable between different users, that can be arranged on most modern operating systems (including Windows); but the computer can still treat accounts separately. The sentient robot part isn't so far from the truth. This actually happened to me this morning - cron, which was incorrectly running in a user-account on spacenimur was sending email to nwmoussa (me, the human), when it should have been sending it to the ntp-daemon - just this morning! Unfortunately, I didn't know what to do with that email and it was written in a language that only the NTP-daemon would speak, and NTP-daemon wasn't getting the update requests, so disaster ensued and the system-clock was way out of whack! Point is, even automated tasks that are being done FOR you should properly be run under a separate login authorization - this is best practice for security. Nimur (talk) 21:13, 21 August 2009 (UTC)

Controlling google results

I am in desperate need of help. Someone posted my full name and other identifying information in the comments section of a video on xtube, an amateur pornography site. I have no idea why this happened. I got the site to remove my name and information, and after a lot of difficulty I got google to remove the page from a search for my name.

It is now 4 months later, and the same page suddenly started appearing on my results again, with the comment visible, even though when you actually go to the page the comment is gone. I don't know if this is a cached page or what. I don't know anything about search technology. I requested that google remove the result again but they keep denying the request. I contacted xtube but haven't gotten any response.

Anyone with a knowledge of search technology, please tell me what steps I can take or who I can contact to fix this. I can't even figure out why it is happening, so if you can tell me that I would also be grateful. Isn't it ridiculous that wikipedia is the only place I am pretty sure someone might respond to this question?

Thank you! Gohome00 (talk) 17:11, 21 August 2009 (UTC)

If by "my results" you mean "the page google returns when I search for my name", then the problem is still Google - they've not removed it from all of the places they had it, and you need to yell at them again. Unfortunately tech companies are particularly bad at responding to electronic communications, so it might be time to write to their legal department at their HQ in Mountain View, California. -- Finlay McWalter • Talk 18:06, 21 August 2009 (UTC)

I would like to not do that but I understand I might have to.... thanks!Gohome00 (talk) 20:50, 21 August 2009 (UTC)

Google can help remove personally identifiable information. There is a link to the website removal request information. Ironically, you must have a Google Account, and submit information to them, if you wish them to remove information about you. [1]. Nimur (talk) 20:20, 21 August 2009 (UTC)

I tried this 3 times. It's an automated form, and if there is some part of the request that doesn't work, all you get is a message that says "denied" with a link to a list of vague possible explanations, none of which apply to me. That's why I'm so frustrated. Gohome00 (talk) 20:50, 21 August 2009 (UTC)

Google books - copyright

Can someone help me out on this google books/copyright thing - I'm a bit confused. The impression I've got is that google has scanned (and used optical recognition to convert to 'text') books; some of which are still under copyright, and then offered them as electronic versions, but doesn't make any effort to observe any original copyright rights due to the author(s)unless they come along and claim them in which case it gives them $60. Have I got this wrong - or does something happen to the copyright when the book is scanned. eg from the Google Book Search article

publishers maintain that Google has no right to copy full text of books with copyrights and save them, in large amounts, into its own database

ok on the surface it seems a no-brainer to me that what is being described is obvious copyright violation (without a prior agreement). What's going on here? Thanks.83.100.250.79 (talk) 20:17, 21 August 2009 (UTC)

Yes, it was blatant and obvious copyright violation; the result was that Google Books stalled for nearly two years in a pitched legal battle (first in 2005), and again in a new legal fight which was recently resolved out of court (despite much continued opposition). After a re-think of the business model, Google Books now partners with advertisers, publishers, and online booksellers; they offer their service as a sort of "advertisement." Nimur (talk) 20:23, 21 August 2009 (UTC)

Any point asking "what were they thinking" (also perhaps even what happened to "do no evil" - hardly saintly anyway)

Can I ask for a second opinion (not that I don't trust your word Nimur - it's just that it seems so blatant as to be almost unbelievable) - is there no explanation for it (excluding economic boldness?) 83.100.250.79 (talk) 20:49, 21 August 2009 (UTC)

My curmudgeonly answer is that a lot of people at google really think that they are changing the world and it is their responsibility to drag everyone else along with them, even at the expense of currently accepted notions of copyright, privacy, etc. Gohome00 (talk) 20:53, 21 August 2009 (UTC)

A bit like WP:BOLD perhaps. Nobody is suggesting a legal excuse though - not even tentative ????83.100.250.79 (talk) 20:57, 21 August 2009 (UTC)

Presumably, one of the things rethought is that you will find sets of a few consecuetive pages missing from the scanned books. I have on occasion found the google books service useful, but of course the bit you want to read is always in the missing bit. Astronaut (talk) 21:37, 21 August 2009 (UTC)

If you look at the McGraw-Hill complaint found here (near the bottom) you can see some screenshots of Google Print as it existed at the time of the complaint. Public domain books were viewable in full. Copyrighted books were only viewable as a few lines surrounding the highlighted search terms (like the current fragment view), unless the publisher had explicitly authorized Google to show more. It's no different from their web search service, which also shows short excepts from each returned page with search terms in context, again without the explicit permission of the page owner in most cases. They weren't serving the full text of copyrighted works. It wasn't blatant and obvious infringement. The copyright code is very unclear regarding this sort of thing, to everyone's detriment. The result of the Authors Guild settlement was an increase in the number of full pages that are viewable, not a decrease, as far as I know. -- BenRG (talk) 00:47, 22 August 2009 (UTC)

Someone said "what happened to 'don't be evil?'" To me it sounds like Google Books did what they did staying true to the mantra. How evil would it be to withhold vast quantities of knowledge from being absolutely free to the human population? Pretty fuckin evil if you ask me. Mac Davis (talk) 01:52, 22 August 2009 (UTC)

I think you should ask authors who depend for their livelihood on their material NOT being free what they think about who is "evil" in this situation. As far as I'm concerned, unilaterally deciding that they know best IS being evil, no matter what the specifics of each situation and initiative may be. Gohome00 (talk) 17:25, 22 August 2009 (UTC)

"Don't Be Evil" is a useless slogan if evil is not well defined. Evil has a huge span of contradictory interpretations. Specifically in the case you mentioned, not everybody agrees that all information should be free - just because you and I think it should be does not make the opposite "evil." I can think of at least a few cases where information privacy might outweigh the "goodness" of information proliferation. Take a look at the unfortunate situation brought up by the previous question - this isn't a matter of copyright, but proliferating personal information, in the guise of "free expression", doesn't sound too good to me. Nimur (talk) 07:56, 22 August 2009 (UTC)

Thanks. I got the impression that they were serving up entire copyrighted books to everyone, rather than just indexing them (and displaying parts in search results). That's more understandable.83.100.250.79 (talk) 09:41, 22 August 2009 (UTC)

Resolved

Blender Fade

How can I fade objects in Blender? --81.227.67.213 (talk) 21:52, 21 August 2009 (UTC)

This was a mystery to me as well, but I experimented a bit and managed it. (I assume you mean "fade over time" rather than just making an object transparent, or transparent at one end.)

My first step was to look at the IPO Curve Editor window. On the right of this window are a list of variables you can vary over time. One of these, helpfully, is "ColA", which is to say the alpha transparency of the object's colour. To insert a keyframe, you go to the 3D view, choose a frame from the timeline and press I. The menu that pops up does not have an alpha option (or red, green or blue, for that matter). So I picked a different option, because I just wanted to get a curve of some kind into the IPO window. You should know that curves in this window can be manipulated like an object's mesh - you can press tab and select and move the vertexes around in the same way (though not in the Z dimension, since there isn't one). I therefore copied my curve to the buffer (press the icon with an arrow pointing downwards towards an orange area - apparently buffers are orange, don't ask me), deleted it (delete key), clicked on ColA on the right hand side, and pasted it. I adjusted it so that over the desired time period it went from 1 to 0, and set the interpolation to linear.

This didn't work at all. It didn't vary the object's alpha. Eventually I found out what it was varying, which was another alpha variable located on a panel called "draw" among the "object" buttons in the buttons window. Just above this is a patch of red colour: the object apparently has a secondary colour as well as a secondary alpha. The tooltip says that these are "used when faces have the ObCol mode enabled". I found that button, ObCol, among the object's material buttons, just above the sliders where you set its ordinary colour. Clicking it made the alpha fade work (and also turned it red). Hope this is what you wanted. 81.131.36.121 (talk) 01:02, 28 August 2009 (UTC)