Chris Blanc: Play and Projects: Blog
Archive for the ‘Web development’ Category
Saturday, December 13th, 2008
Google this week admitted that its staff will pick and choose what appears in its search results. It’s a historic statement – and nobody has yet grasped its significance.
Not so very long ago, Google disclaimed responsibility for its search results by explaining that these were chosen by a computer algorithm.
A few years ago, Google’s apparently unimpeachable objectivity got some people very excited, and technology utopians began to herald Google as the conduit for a new form of democracy. Google was only too pleased to encourage this view. It explained that its algorithm “relies on the uniquely democratic nature of the web.” ^
We like things that operate invisibly because then we assume they’re fair because no other single person could be gaming the results. We trust nature, the Invisible Hand, economics, democracy etc because they are in the hands of the mass and not an individual.
Yahoo! took one look at the web and realized that, unless AIs got a lot better, human beings would be needed to (1) understand what the person behind a search query is looking for and (2) separate the wheat from the chaff. This applies to complex queries; someone looking for “Britney Spears” does fine with auto-democratic results, but someone looking for “doors of perception” needs to be redirected to Aldous Huxley and not “Perception-Enhanced Doors, Inc” of Ocala, FL.
With Yahoo! having, through the type of bad leadership endemic to programmers who are accustomed to each object having a linear function, gone belly-up for all practical purposes, Google is closing the gap: now the machine does the hard work of compiling possible links, and people vote them up. It’s like Wikipedia, Slashdot, Reddit, or Fark… but, using the lessons of Wikipedia, Google has appointed an editorial staff so the results don’t get hijacked and mutilated.
It’s an intelligent move, and that required Google to overcome our innate modern fear of The Other Guy and start selecting people who can lead and putting them to good use: making sure Aunt Hilda is able to find accurate results, sans spam and idiocy, every time. But somewhere in Jerry Yang’s mind he should be registering that he was right all along, and something else caused Yahoo! to curl up and die.
Posted in Web development | No Comments »
Monday, December 17th, 2007
At the end of The Bourne Identity, Jason Bourne is at the top of a staircase with a bad guy below who has the drop on him. His solution is to kick the corpse of another bad guy through the stair rail and ride it down, implausibly plugging the (living) bad guy as he falls. Google does the same thing quite frequently.
When they wanted to take out Microsoft Office, they encouraged many companies to develop online apps that looked like Microsoft Office, and then bought the best after Microsoft had snapped up several others that are now worthless iDung.
When they wanted to take out Microsoft Internet Explorer, they pumped money and developers into Mozilla Firefox, making it as corporate of a project as IE. Result? Many people use it, believing it to be a real alternative, while Google slowly slackens its support into the background.
Finally, Wikipedia: Google needed a way to provide some kind of standard result for any search query, because too many people were spamming. So they encouraged wikipedia, knowing that its content would eventually get out of hand.
Now, they’ve introduced Google Knol, which is a wikipedia clone — except that it’s hybridized with a group blog, and is only open to select contributors. Thinking of Associated Content or Reddit? Yeah, me too.
It’s a good way of acknowledging what Wikipedia tries desperately not to let the world know. Most wikipedia articles are written by relatively few people, maybe 2% of the contributing audience. They are augmented by another 10%-20% of the people there. The rest of the people on Wikipedia perform really obvious monkey tasks like plagiarizing websites that are expert in their area, so the Wikipedia page appears above them in search results. This was basically a giant web real estate grab.
With Knol, Google is starting where Wikipedia left off. Google has no problem admitting its elitism, and that it wants proven writers (probably power pro-bloggers) to write for knol. It doesn’t want the infighting that makes Wikipedia unreliable. It wants the hard content on technology, science, and social topics. That’s what’ll make it money.
Another smart strategy from Google, but maybe a bridge too far. I would further hybridize Knol with social networks to make sure that communities formed around any specialized type of knowledge, and they collectively produced it, including the makers of those original websites that Wikipedia plagiarizes. These are the people who really know what they’re doing and they deserve to have credit go to their websites, and then they will support the Knol concept.
This is all part of the web correcting itself. It became radically de-centralized, and then people saw that web pages were lost each time some guy who had the best web page on some topic had a personal crisis and let his blog registration lapse. So that knowledge, and more importantly, the way it is presented (knowledge ordering context) is moving up the chain toward centralized sites.
It’s not that much of a change from the past. Imagine if dmoz.org or dir.yahoo.com had archived sites into a standard template, instead of simply linking to them. That’s how far the web has come in 13 years :)
Posted in Web development | No Comments »
Monday, December 17th, 2007
Recently Google, the “new Microsoft” in both its power and good/evil split inherent to larger corporations, introduced what’s been brewing for some time: primitive social networking style Profiles. Naturally, they’re going to bump any information found in these, so you want one (check the Chris Blanc profile(tm) for an example).
Simple steps to get one:
1. Log in to any Google service, like, say, Gmail or iGoogle.
2. Go to Google Maps
3. Next to your login name at the top of the screen will be a My Profile link. Click.
I’d appreciate it if you linked back to this blog or any of the good causes found in the left-hand links list. Send me an email if you do and I’ll send back something vapid yet sweet, twisted yet wholesome, and perhaps otherwise perplexingly convoluted language.
Posted in Meme Trafficking, Web development | No Comments »
Tuesday, September 4th, 2007
I resisted the blog for many years. I think all things in life get branded by how they appear to most people who aren’t totally brain-dead. These appearances define their archetype. These archetypes get manipulated by your favorite TV shows and comedians as well, but some things just appear right. One of those is that the blog is for those who want to bemoan their problems, decide not to solve those problems, and make you suffer through reading a steady stream of self-pity. It was this cause that made me deny the blog, but I gave in, finally. Information might be the last vital commodity on earth, and when I encounter something that is truly informative, I think it belongs here.
On another topic, I figured I’d mention that we host with the much maligned Dreamhost. They’ve grown a lot over the last two years, and run into some problems, but they fix them rapidly and transparently, which means they tell you what’s going on and how much they know it’s going to take to fix it. I’ve had a number of webhosts over the years, and like this one the best.
For professional clients, depending on traffic, I either get them hosting with a smaller ISP or rent rackspace and install a quality operating system and web server. If your traffic is enough to warrant this, it can be a great way to host!
Posted in Web development | No Comments »
Tuesday, September 4th, 2007
I feel like a heel for mentioning it in passing in previous posts, and not elucidating. You probably hear a lot about Windows Vista, after all, and most of it is negative in that familiar “David versus Goliath” style of logic that many people seem to adore. After all, Microsoft is a big corporation and we’re little guys, so we should hate them. Never mind that the same socio-political system that sustains us created them. This blogpost attemps to clear the air.
 Before we even talk about Vista we should mention a truism of technology. It is unwise to be an earlier adopter of any new hardware or software. The first year a car is on the market, it will have problems. The early versions of a new DVD player are going to need firmware upgrades and possibly hardware swaps. Any version of software ending in “.0″ will be followed by a .1,.2 and .3 which might actually work for everyday tasks without crashing. It’s nearly impossible to predict all of the situations a complex piece of technology will encounter, so expect early adopters to crash test it and some months later to have something you can use (I say nearly impossible, because it is possible to get very close, and I’ve done it, and intend to do it again because the gratitude of your users is priceless).
For this reason, I’m not even going to think about installing Vista for 2 years, and will probably only do so when I have a need to upgrade hardware. I like keeping the mound of landfill generated by my life low, and so I’m not prone to buying things I don’t need. A well-built desktop computer will last ten years, although you’ll want to replace hard drives every four years. At the ten year point, it might still be running, but it will be so hard to find parts it is wisest to move on. I anticipate that at some point two years from now, I’ll pop out the motherboard on one of my clones and put in a new one, possibly used equipment. It’s why I like PC clones: you can fix parts instead of buying an overpriced, decorated, physically unreliable piece of technology. Before you consider pre-fab, realize that the best green computing you’ll get is something that doesn’t require you to buy a whole ensemble every three years.
Two years from now Vista will have undergone some changes. First they’ll fix the bugs. Then they’re going to try to eliminate all the duplicate code they can, and trim down its estimated 50 million lines of code. You do this by finding parallels in function between bits of code, and inserting abstraction layers so they can be shared. Do too much of this, and all of your processing power goes to unhelpful abstractions. Do just enough and the system balances faster execution with less variance, so is far more stable. One reason UNIX is so stable is that it has grown organically from very few functions and so is heavy on the sharing of building blocks, which can then be made stable over two-year cycles.
At that point, you will be able to see what Vista really offers. As with many things, to understand what it does well you must understand what it is not doing, and to do that, it helps to have a refresher on operating systems. Because they’re software, operating systems are dependent on hardware, and the evolution of hardware has driven the development of new operating systems. A computer that saves data to punchcards, has 2K of RAM and runs at .5 mhz does not need an operating system other than the ability to send data to its cardpuncher. A computer running at 20mhz with 2MB of RAM can take advantage of multi-tasking, protected spaces, optimized disk use and so on. We’re a few stages past that now.
The operating systems currently on the market are based on the hardware of the 1970s and 1980s. This hardware is a relatively fast processor, a bunch of cards thrown in for interfacing to peripherals, and a CRT monitor. There is little collaboration between coprocessors, and no consideration of multi-tasking. The graphics cards are primitive devices that receive instructions about where to place pixels and then keep track of those pixels. The disk drive was often treated as a random-access punchcard with large capacity.
Since that time, computer hardware has improved in every way, but no one has quite managed to stitch it all together with a new theory of how it works. Oddly, this will probably happen after Vista becomes commonplace and drives the next round of upgrades. In my experience, history oscillates between concepts (shades of Hegel, without the illusion of forward motion) in order to advance itself like a ladder toward higher degrees of complexity. Thinking back to the days of mainframe computing and its descendents on the desktop, including the Amiga and NeXT, I think the future is going to be a time where the main processor spends its time managing resources as much as computing. It will be aided by coprocessors for graphics, I/O and specialty calculations, and will with the rise of the hyperthreaded processor become more like the “cell” computing paradigm.
In cell computing, each computer is made of up many processors, each of which is like a complete networked computer. The advantage is that complex tasks can be tackled in parallel. For simple uses, there is no advantage. This bypasses the older model, in which a single powerful processor was king and was surrounded by subservient but brick-stupid helpers who handled the rote chores of graphics, I/O, sound and so on. Vista addresses that need as well.
Speaking of history, there’s a second need Vista addresses, which is backward compatibility. PCs at this point can run just about any program written since 1983. This is an awesome advantage when retrieving old data from proprietary formats, or playing retro video games, but it is also a liability. There are thousands of obsolete instruction pathways the operating system must support, and it must limit itself to older paradigms. My guess is that Vista will include a virtualization engine so that when you need to run 1983-2008 software, it will do so in a “virtual PC” running a small version of Windows XP. That way everything old will work flawlessly, but so will the new.
Now that you’ve got all of that crammed into your head, you can look at the advantages of Windows Vista:
Threading. It takes advantage of the new processors’ ability to tear a task into parallel silos, run through them at high speed, and generate a result. It’s not the same as multitasking, but it’s like multitasking by dividing a program into six parallel functions that run at the same time. As you might guess, a big requirement is that the parallel pieces can communicate in real-time without getting hacked or lost. Vista’s threading model makes a number of positive changes. For more information, read Inside the Vista Kernel.
Graphics. In the old days, you stored data in pixel form, like you’d draw it on the screen. Now that screens come in a variety of sizes and shapes, and our graphics cards contain processors nearly as powerful as the computer itself, we don’t need that literal model. Vista uses an abstraction layer to store graphics as vectors, or abstract shapes, and then translates into pixels at the hardware level. For more information, read Vista is a hardware beast.
Interface, functional. Vista acknowledges that the old operating system was file-based as a way to manage that newfangled thing called a “hard disk.” People now want a file system to be an information manager that actively keeps track of their contacts, junk, links, files, IMs, and so forth. Vista is based around the idea of indexing and searching data. For more information, read Vista Beta 1 versus Tiger.
Interface, visual. Windows XP was a visual update to the same basic interface we’ve been using since Windows 95. It’s time for a change, especially one that takes advantage of the on-card effects offered by newer video cards.
Programming interface. Years of compatibility, work-arounds for hardware and software, and shortcuts have gone into the Windows 32 Application Programming Interface (API), which is how programmers make their programs cooperate with the operating system. They put in some new stuff, and for a change, the names are consistent to make them easy to remember. For more information, read Changes to the API in Windows Vista.
Meta data. Filenames and file extensions don’t tell the whole story. If it’s an MP3, who is the composer? If it’s a drawing, what is needed to view it, and is it vector or bitmap? Meta-data is how search engines will sort files in the future as well, and is one of the great promises of XML as a data categorization and storage mechanism. For more information, read Windows Vista Beta1 Vs OS X Tiger.
Speed improvements. Better caching, better prioritization of multitasking, and a brand new driver setup to improve performance, eventually. This won’t hit for awhile as there’s still too much code in Windows Vista for it to be fast. Expect improvements from MS on this front over the next 18 months. For more information, read Bit-Tech MS Vista Review and Windows Vista Speed Tweaks.
GUI improvements. There’s a “Welcome Center” that groups common functions, and makes much more sense than the procession of icons previously used, and a NeXT-style sidebar, and an improved desktop and more readable fonts. Of these, the improved Welcome Center is what will benefit users the most. For more information, read What Windows Vista Does for Me.
Better built-in applications. Long ago, computer makers tried to throw as much software at the new user as possible, so the user had an expectation of having a working machine when they got it home. By working I mean “can do something real-world useful with it.” Then we got into the mentality of shipping white boxes with an OS, and hope the users know someone down the block who can burn them some CDs. New DVD, CD, video and audio authoring software come with Vista and it’s a good thing for the end users like Mom and Dad who just want to make some vacation videos. For more information, read What Vista Gets Right.
Increasing reliability. It has nowhere to go but up, although I think the problems with Vista are overstated while problems with other operating systems are understated, simply because the public loves an underdog. 50 million lines of code means at least 500,000 bugfixes. For more information, read Vista Developers Blog.
Important Note
Today’s conventional wisdom, based on more than a year’s worth of relentless negative publicity, says Vista is hopelessly broken. In fact, my experience says the exact opposite is true. I proved the point in the first installment of this series, where I restored a sluggish $2500 Sony Vaio notebook to peak performance in a few hours. And I think anyone with a modicum of PC smarts can do the same. ^
There are other details, but these are the major reasons Windows Vista is important (check the Windows Vista Team Blog). I’m not going to delve into the Linux-vs-Microsoft-vs-Macintosh thing here except to say that when Vista is stable, it will be a more advanced model than either one (Windows NT, as a descendent of DEC’s VMS, was already more advanced in theory than Linux, but it was hard to see past all the bloatware).
I’m not really pro-Microsoft or anti-Microsoft. They hire some brilliant people and they make their mistakes like any large corporation does, and corporations are the basis of modern capitalism, so Microsoft hatin’ is really another topic. I give them credit for doing what the market wasn’t able to do before them, which was produce a unified standard for which programmers can write software with the expectation that just about anyone can use it, and for making Windows XP an operating system that runs on almost any hardware and is very stable. Beyond that, I’ve got no investment in this issue.
I do think that, as much as corporations unleash Fear, Uncertainty and Doubt (FUD) against each other’s products, the computer industry has a lot of bitter people who will unleash FUD against any product from a company that’s making it, or making it more than they are. This was particularly bad after the dot-com bust, when people refused to use Google or other success stories because of their own lack of success. Vista is here, and like everything in life, it’s a mixed bag, but I wanted to tell the side of the story that shows some things to look forward to, regardless of what FUD from any “side” of the story tells you to think.
Posted in Information Technology, Web development | 3 Comments »
Tuesday, September 4th, 2007
We all got the message over the last two years: stop using Internet Explorer, because its ActiveX control friendliness is prone to security errors.
Most people then switched to Mozilla Firefox.
What’s amusing is that these browsers both descend from early-1990s engines. IE is a grandchild of Spyglass, and Firefox is the great-grandchild of NCSA Mosaic and the grandchild of Netscape Navigator.
There is one browser authored anew, with new standards, and the explicit goal of being small and fast and solid.
http://cybernetnews.com/2007/09/03/cybernotes-exclusive-opera-95-features-video/
Check out the new Opera 9.5. I’ve been using Opera since 8-series and have been impressed to the point where it is my preferred browser. Its userbase also does not have the fanatical, ugly, supercillious chip on its shoulder that Firefox users often do.
Impressive features:
Search every site you’ve visited
Faster page rendering than any other browser
Full standards support
Restore closed windows
Synchronize bookmarks remotely
Quite impressive! You can download it here:
http://snapshot.opera.com/
The browser wars that raged first in the 1990s, until Internet Explorer won, and were revived when its security problems drove people to Mozilla (in the same way Microsoft’s continued inability to secure older installations of Windows XP, and the slow process of perfecting Windows Vista) are now in turn revived, because Opera has upped the ante. For the first time since 1997, the browser is substantially changing in function and becoming closer to the general information peruser and manager many of us have always thought it would be.
Posted in Web development | No Comments »
Thursday, August 30th, 2007
You’re on some website, and you’re sure that the text you’re looking for is somewhere in the mess of pages. Problem: the site doesn’t have a search feature.
Bookmark the following link (right click, “Add to Favorites”) and you can search any page by going to that page, and then clicking the bookmark you make of the link below. The script will pop up a small window into which you can type your search term, and then see Google’s results for that page.
Search Current Site
Posted in Web development | No Comments »
Saturday, July 7th, 2007
Although people know me as a writer of fiction in my personal life, my day job is fixing computers, so when someone I know needs work done, I often get the call. Recently someone I used to work with as a computer consultant called me up to help install a home PC for a sibling in town. Because I’m nice and/or stupid, I did it.
The experience was not bad at all, since the user was kind and understanding. What stuck in my head however is how not ready for use the average Windows installation is. They try to get the user going with some kind of video-based training, but that isn’t going to work for most people who find it horribly tedious and would rather play. My approach is different: I clear out the unnecessary, organize the computer appropriately, and then let ‘em rip so they can learn by playing.
Here’s a list of important steps:
1. Disable all crapware, even if it seems useful. Rip the installed programs list down to the bare minimum.
2. Update all drivers. I’ve rarely had to roll back.
3. Set the system swap file to be twice the size of the memory installed in the machine.
4. Add a firewall, install FireFox or Safari or Opera to replace Internet Explorer, and remove virus scanners.
5. Add Pegasus Mail, Thunderbird or Opera mail to replace outlook, and disable HTML mail and .exe, .vbs and .js attachments.
6. Clear all extra icons off the desktop. They need My Computer, My Documents and the Trash and that’s it.
7. Uncheck all protocols except TCP/IP in the network configuration.
8. Set up a user account with Administrator rights, but keep the Administrator account in case they need bailing out.
9. Set up a guest account for drunken friends, idiots and clueless family members to use. You can set the password to TOIDI or NOROM to really fool ‘em.
10. Run HijackThis and remove all unnecessary start up actions.
11. Go through your task manager, using this guide or one like it, and remove unnecessary programs.
This will take you the better part of three hours, but if you do this and then backup the system, you will have set your friends on the path toward sanity instead of confusion. They’ll thank you someday.
Originally written on Slashdot
Posted in Web development | No Comments »
Monday, May 14th, 2007
Jakob Nielsen (and you should listen to this man) tore Web 2.0 a new one today, pointing out that the glam and flash of nifty interfaces has obscured the honest truth: that people using the web don’t use it for its own sake, but use it as a tool for some other purpose, absent that relatively small but vociferous group of basement-dwellers who inhabit the web like an identity.
I’ve been saying this since 1994 or so, and it has helped me advise clients and friends well. You want to make something people can use for a function you have to offer. If you have no function, either invent one (put out content germaine to your topic) or keep your web presence at the level of a muted hybrid between advertising and a phone book listing. Some people have initially resisted this and later found it meaningful.
This brings us to the question of philosophy in the indefinite sense, meaning philosophy as a language in the usage that gives us the idea of “a philosophy of” each topic. Is there a philosophy of the web? Well, first there is a language for discussing the web in the sense that philosophy excels at, which is describing the invisible but structural connections between ideas. Anyone can describe a widget. A theory of widgets that connects use to design? Rare.
After we stumble around with more verbiage about whether a philosophy is the philosophy, and the many types of philosophy and their possible appeals, we come to an end of the line of thinking, as must always happen when we leave our minds to make things work in whatever shadow of The Real World we manage to inhabit. The web is made to be used. Similarly, in my view, philosophy peaked with the ancient Greeks because all of their theory aimed toward what we might idealize as “common sense,” but really was plain arguments for making life better through systems of thought.
There will undoubtedly be a seesawing argument over Web 2.0 in the coming days, in which one side will say Web 2.0 is style over substance, and the other side will claim it is the second coming of Christ. All I can say is that for me, what matters is connecting a target audience to a function that makes their lives better. If Web 2.0 helps, as it does with Gmail, I’m all for it. But if an earlier technology works better, I won’t shirk from that either.
Recently I switched back to 1970s technologies for email. I use Mutt on a UNIX box, without the benefits of GUI or really any extended function, and I like it because it requires deliberate usage. Keeping track of friend’s emails and aliasing them is an active process. Sending an email requires thinking about each step deliberately. And there’s no glam and flash in the background. Like design itself, usage must match form and function some master science of philosophy that describes the context of usage itself — a thing often known as “life.”
Posted in Technical Communications, Web development | No Comments »
Bolg – The Chris Blanc Weblog is proudly powered by WordPress
© 2010 Chris Blanc
|