Since SmartTags, Microsoft´s new feature that adds extra hyperlinks to web pages based on the content became public, there has been a lot of whining about it. And it´s all hypocritical.
Since SmartTags, Microsoft´s new feature that adds extra hyperlinks to web pages based on the content became public, there has been a lot of whining about it. And it´s all hypocritical.
Since SmartTags was unveiled, there has been a lot of reactions to it.
But what is the rationale for all this bashing of SmartTags? You know, I had thought of something similar a few weeks ago, and it really seems like a nice feature! And even if it is not a nice feature, who cares? No software is perfect, and if it really sucks, it will go away
Ok, so this tool can be abused? Big deal. It happens all the time! Reimplement it. And do it better. If you think MSIE with SmartTags gives useless links, or even damaging links, write software that GIVES BETTER LINKS.
Make Galeon, Mozilla, Konqueror, Opera, whatever, give cooler extra linking. If the users want it to be easy to disable, make it easy to disable. If the user wants the "extra" links to be easily recognizable as not part of the original page, make it so.
And why am I so sure that the whole backlash is simply because it comes from Microsoft? I am sure because linux users have been, for years, using software that modifies the webpages and REMOVES information from webpages without the copyright holder´s agreement.
Where is the outrage of the amateur lawyers? Where is the anger of the web designers? The Slashdot coverage? Nowhere, that is where.
So, stop hypocrisy. If a feature seems useful, don´t condemn it because of the one providing it. If it doesn´t seem useful, just don´t use it, or build a better one.
It is not through whining that free software is built.
, but with every invention or advance in science, medicine or technology, also new disasters, terrors and negative side-effects are invented.
With nuclear power, nuclear bombs were also invented. With airplanes, planecrashes, with antibiotics, resistant viruses and so on and so forth. Smart Tags don't fall into categories this big, but they will have their impact, both positive and negative.
The positive are placing information nearer to the consumer. If I read an artcle about A, I do apprecite extra information about the subject, the author of the article and the magazine the article is published in.
The negative are the possibilities to lock the user into a mini-internet (webpages of the same company or daughter companies), giving misinformation (this is on of the great disasters invented alongside information technology - misinformation and witholding of information) and generally altering it.
Personally, I find the idea of Smart Tags in the current implementation disgusting - I do not believe that a company will ever release information that can cause itself harm. Ever, even if the information is true. When we consider Microsoft that is _everywhere_ we will get an internet (remember, IE is the nr 1. browser) where pro-microsft links are always more available than anti-microsft links. Chances of clicking the first increase proportionally with decrease in the case of latter. Effectively we will be witheld certain information (anti microsoft) while propagating other kind (pro-microsoft). It is just statistics, nothing else really.
And please remember that MS is not "the company that makes the OS & IE. Microsoft has very many interests and probably countless contracts with third parties in other areas (so we will see more links to company x and less to y). And this is not statistics anymore, this is control.
About Junkbuster, I'd add that *I* control what it does. It's not even close to being the default anywhere. *I* choose to install it. It's not like there's a Junkbuster Corporation deciding that they might let one of their "friend" company's ad go through, blocking only their competitors.
It doesn't matter who makes the web browser--Microsoft, Netscape, AOL, the Galeon project or the KDE project. What matters is that somebody with near-universal browser share has decided to edit the entire web. I'd be just as upset if somebody decided to ship Junkbuster with the vast majority of new computers.
I'm not looking forward to the day when Microsoft puts little links to all their website all over my pages.
Sounds similar to that plan Deja had to insert sponsored hyperlinks into archived usenews postings.
In that case, Deja was effectively going to republish derivative works without consent of the authors. It didn't make the modifications (hyperlinks) clear, so they might be attributed to the author, or change the meaning of the original document.
SmartTags seem to do a similar thing, except that the modifications are merged with the original document at the browser, so it is not really publishing the derivative work. People are still going to be pissed off if it isn't clear what parts of the document are modified. Even though MS wouldn't be distributing modified works, I could imagine a court ruling that that is effectively what they are doing.
As far as private use of ad blocking software goes, I don't see anything wrong with that. I could imagine an ISP getting sued for blocking ads for all their customers though (or even worse, substituting their own ads).
How is this different from what Deja[News].com was doing, adding links around "key words" in archived posts? Won't roughly the same horde of irritated people be shouting at them about copyright violation? Or is that what all the articles are about :-)
It is not through whining that free software is built.
Sure, I'm all for better linking ... in a user controlled and standardized fashion !
I would love to see support for better linking support in Netscape, Mozilla, Konqueror, Gtkhtml(2), etc. especially implementation for the XLink specification. I would especially be interested in seeing external link bases support, which allow to build distributed (and hence scalable) annotation systems.
However I doubt I will see this soon in Mozilla or Netscape, and I am afraid I will be able to use them only a few years after it gets available on Windows, maybe I'm pessimistic.
We should recognize that Microsoft made a serious work in IE and we are behind in a number of areas, the fact that MS now tries to turn to its advantage the fact that it's by far the leading browser is not surprizing, it's a for profit company ! It's good to show when they make something nasty, but we should also recognize its merits. When Netscape was the dominant browser they completely ignored or were a pain for standardization, not surprizing isn't it ?
My initial reaction to smart tags was that it was a cool idea. I would really like to be able to turn on some kind of cross-referencing engine whenever I view documents. That could be an incredibly powerful tool for research, and probably even entertainment as well.
The problem that I see with it is not so much that it alters the documents, and takes away editorial control, but what the links will end up being. No one should really be able to claim that they are going to have complete editorial control over a web page....it is going to look different in different browsers, people will run junkbusters on it, etc. If someone runs a client-side cross-referencing agent, that just increases the value of the document, and likely can drive up hits to that website, assuming it is of value. But this last idea, driving traffic to websites, is exactly what Microsoft appears to be exploiting. Rather than doing smart tags solely because it is a good idea, and the best content will be linked, they are mixing advertising with editorial content. That is a really bad thing for journalism.
I have similar thoughts towards UDDI - it is a cool concept, but the implementation is geared towards finding a business, not a service or resource. There are a lot of cool ideas for helpful technologies out there, but the problem is come implementation time, they are done in a way to benefit the implementor as much as possible. It is an understandable decision, but it is narrowly focused, and will do more damage than good in the long run.
I´m afraid you might be falling into a little "Microsoft is evil" prejudice here, unless of course you have used SmartTags and it does drive traffic to Microsoft´s sites.
I would say that even if it did, it would still not be bad, as long as the links go to relevant information (say, the encarta definition if everything?).
BTW: another example of a tool to change the contents of webpages before display: translation software. Who ever complained that babelfish should be illegal? ;-)
Further: Slashdot and company used to give babelfished links, IIRC.
Hasn't *anyone* drawn the line between this weird little "SmartLinks" thing and the "What's Related" feature in many browsers? The only difference is that the smartlinks would be embedded in the document, thus more visible. I'm not saying that either feature is good or bad, just pointing out this correlation that I haven't seen anyone draw.
If Microsoft weren't a monopoly, if IE weren't the by-far domninating browser, then perhaps, the idea could've worked. Then people would've had the choice not to use IE to browse the web.
Now, we'll instead see they use their influence to trick even more people to Microsoft web sites, to read their views on life and software, to strengthen their monopolies.
Yes, we can make mozilla or konqueror use different links, but some 80-90% of all people will get the links supplied from Redmond, or from companies that paid the Redmonders.
The 'related links' side tab was quite a different beast as it was not that intrusive (and we all shut if off rather soon).
Let's say I post a rant on the web complaining about how "Company X" in "Country Y" stole my money and sell substandard goods. This is fair enough since I provide proof that my complaints are justified.
Then someone working for a totally unconnected "Company X" in "Country Z" (where I live) happens to find my website while browsing using IE6. The "Company X" that I have had problems with have no website so I do not produce a link to them, however MS IE kindly adds a link to "Company X" in "Country Z", and so the employee now begins legal action with the intention of suing me for slander.
My point is adding links into documents can change the meaning of them, in some cases dramatically. This is what differentiates it from "What's Related" and similar features. You may also say that I could have added the meta-tags to stop the link being produced, but why should I be dictated to by a company in Redmond, each time they add a feature to their software. I think smart-tags are a stupid idea, but if Microsoft insist on them being included then they should be opt-in, not opt-out.
so. you want to view web pages of the fourth reich. however, your ASP has decided that content written by nazis is Bad. Therefore, it will be removed.
now, this has two effects. one, you are not offended by misleading content. two, by not seeing content that you may think to be offensive - e.g. a 'historical document' that denies that the holocaust that killed 6 million jewish people, persecuted romanians, gypsies and basically anyone that didn't fit the 'aryan ideal', - you may not ever get so utterly incensed and upset at this mis-information that you decide NOT to write a letter to your MP, NOT to anonymously SPAM their site with such anti-naziist comments that their site crashes and is taken off the air.
i know where this 'smarttagging' will lead. censorship. now, i know a lot of americans would be _more_ than happy never to see certain things ever cross their eyes or ears (e.g. the bible-bashing belt that cause newspapers like the AJC to have to write things like 'The World, believed by some to have come into existence with an event describe as the Big Bang, ..." so that they don't get hammered by letters telling them that _of course_ the world was created in seven days.)
these people, i am sure, would ADORE smart-tagging censorship. right up until the day that a terrorist posts a letter-bomb through their door because they happened to belong to a particular group or neighbourhood, and they never saw the news reports that their neighbourhood was being targetted by anti-whatever terrorists because they decided, or someone decided for them, that they didn't want to hear about terrorism?
uh.... GOOD! bye bye.
am i scaring people enough, yet?
if not, let me know, and i can think of a few more nasty scenarios.
all that having been said, using content-modification for the _right_ purposes, e.g. for generation of different site templates using the same content, is technically sound, and a good idea. design one site, then use a standard WAP-template, a standard-netscape-template , a standard lynx(text)-template, GREAT idea. use libxslt to implement it.
customising how you see something? great idea!
removing content by having filter-processing at an ISP level? baaad, BAAAD idea.
as a responsible technical community, we should consider how it may be possible to add digital signatures to content, and how this can be added to browsers or to trusted proxies that a standard browser can reference, in order to ensure that content signed by a web site is NOT modified en-route from the server to the screen.
does that HTTP-auth/digest HMAC_MD5 rfc sound like it might do? isn't that already an established mechanism? [and does anyone know what i am referring to, that has rfc references they can post here? :) ]
Ok, let me see if I understood you correctly: you are saying we should forbid SmartTags because it will cause censorship in the future?
Excuse me if I say that makes no sense and I hope I just completely misunderstood you ;-)
I say that makes no sense (just in case) because:
a) You can not forbid something because of a possible evil side effect that has not happened yet. If you do, the internet should have been forbidden in 1971 because it would cause spam.
b) Forbidding smarttags means " let's censor this little thing now so we are not forced to censor a lot later". I think that reasoning causes a slippery slope that in the end invites widespread censorship.
With smart tags you are simply not browsing the web any more, but a modified web. Not saying anything is equivelent to accepting the use of this type of technology, and therefore accepting a modified web. While something like smart tags might seem harmless, this will lead the way to more and more content modifying techonology. Taken to it's logical conclusion the web you will browse will be the one intended by the makers of your browser.
A similar thing is happening in the legal area, where Goverments such as France are trying to remove content such as Nazi memorabilia. Society has said what the Nazi's did was bad, so the logical thing to do is to remove such bad influences on the web right? I would say, alowing a government to behave in that way can only pave the way the more such behavior, which is censorship of the web.
Even on a more basic level, adding such links changes the semantics of a document, however subtly. It is esentially taking a document and interpreting it for another person. In a document you can stress importance of something by empashizing, bolding, or linking text. It's sort of like being a librarian and giving someone a book with markings of the parts you thought were important, and perhaps even writing in the margin pointing where to go for furthure study. On the other hand, say you like a book, you might ask the librarian what books are related or where you can go for futher reference. SmartTags is the equivelant of the over intrusive librarian, who goes out of their way to direct what you read and what you should find important.
What makes this issue even more important, as others have pointed out, is Microsoft's overwhelming browser share. If this was a small free software project, I'm sure most could care less about what they did with their browser. The fact that some 90% of internet users will unknowingly view a modified web is cause for concern. It our job as people on forefront of technology to call attetion to and to bring about a debate on things with such far reaching affects.
for those who prefer to make up their own mind, there's plenty of documentation over at the microsoft office site. start here
In France, you are considered responsible for what you publish (in a newspaper, a journal or on the web). Certain things are considered illegal to publish. One of these is negationism: openly advocating negationism in any publication will bring you to court.
You could discuss whether or not it is a good idea to have certain things be illegal (for exemple, calomnia) but the fact is that the web is not a place where no law exists. At least, in france, it is censored as much as any other publishing mean. I happen to find this good but this is another problem.
I think there would be less controversy if we knew Microsoft would not provide default glosses of its own. Yes, Yahoo, MSNBC, Wired, etc would all put "Add us to your smarttags list, click here!" in their banner ads, but people would only add 'em if they saw direct value. I mean, if Opera or iCab added this feature, we'd all love it, because we know they don't have a non-browser agenda to pursue.
...if the smart tags were controlled by the user. I'd love to add my own annotations to pages I visit. Bookmarks, as implemented in most browsers, are pretty bad.
But annotation prepared by some marketing guy at MS!? No thanks!
First let me start by ignoring all moral implications of changing someone else content for a second ...
If you actually go look at how this stuff is implemented, it would be fairly easy to down load a set of XML files that give a "smart pointer" perspective that more suites your style and the community you feel part of. For example, I could have any time I saw "GPL" take me to fsf.org and any time I saw "Voice over IP" it could take me to dial911anddie.com. Both sites I will be somewhat surprised if the default Microsoft smart tags take you too :-)
My point is, if you read the Microsoft documentation, we might have more control over what happens than most postings about this topic lead you to believe.
Well I am a webmaster, and you are damn right I am against this.
For one simple reason. If I write "linux" on my website, I don't want my colleagues, who read my site in IE, having it linked by their browser to a Microsoft article about how crap Linux is.
The point is that this turns the whole web into a huge advert, for microsoft and anyone who pays microsoft to do it. I refuse to have my website advertise companies or products I do not condone or even know about.
If I want to link to articles on msn.com, I will do it.
I happened to read that documentation before before I noticed the uproar. Trust me, if you do things in that order, most articles on this topic seem to be written by complete morons.
After all, this feature isn't that different from something most browsers already offer -- the ability to select a text snippet with the mouse, and execute an action on that text. The only difference is that I can get the browser to find such text for me, and tell me what will happen before I execute the action.
And since I happen to believe in personal freedom, whether or not I chose to use such a tool on my computer is my business. If I like it, I'll switch it on. If I don't like it, I won't use it. It's my choice, not the choice of some brain-dead pundit or control-freak publisher.
(BTW, according to recent reports, the IE6 will ship with smart tags switched off, and with no standard tags installed. The user has full control, as it should be.)
Suppose I invent a phone that plays advertisements occasionally during calls, and suppose I use all my powers to get people to use these phones.
Is that okay? Sure.
Unless I'm AT&T prior to the breakup.
There are things that are okay in general that aren't okay for monopolies.
Hypocricy has nothing to do with it. This is about control of the content of the web. I wouldn't like this if it were included in Opera, in Mozilla, or in Mnemonic. This is Just Plain Bad, in implementation.
The idea is okay. Not great, but okay. The idea of providing links to various phrases and words contained in a document is nothing new. There was that lame commercial a few months back bragging about some stupid software that did the same thing.... only differently.
But, back to control of content. There shouldn't be a new tag to turn the feature off; there should be a tag to turn the feature *on*. As publisher of a web document, don't I deserve control of the content of my page? And don't tell me hyperlinking isn't content, because it is. That is one of prejudices, I suppose.
The big question of hypocricy is this: would Microsoft itself complain if Netscape had introduced this feature years ago, and every mention of MS-Windows on Microsoft's own pages linked to microsoftsucks.com? This is essentially what MS is doing. Can you imagine the links that the word "GPL" will give?
This has nothing to do with innovation, or even good ideas. It has to do with content control. If there was a tag to turn it on, rather than off, I suspect I would feel differently. But this is an opt-out, rather than an opt-in, feature.
Let me ask this: do you prefer opt-out spam, or opt-in spam?
Hey, if you want unfettered control of your content, you just have to forbid:
* access with any browser you have not tested and approved * access without the client-side scripting you want to use * all browsers with nanny software enabled * access from sites known to display modified versions of the content.
* altavista * google * anonymizer.com * junkbuster * pretty much any proxy that does content-checking * all nanny sites
And, why not, all proxies, since caching may cause your page to be outdated, and you surely do not want that.
Somehow, I don't feel like you are asking or willing for this, but only opposed to MS doing smartlinks.
Of course, if you ARE asking for this, and you have taken measures to do it, I would love to get a URL to the site where you are implementing it.
I would believe your site is a crass exclutionary page which goes against every promise of the web, but that is just me, and you would be coherent.
I prefer no spam, thank you, either opt-in or opt-out.
Most web browsers are designed to present the *content* of my website unaltered-- you are mixing up presentation and content. I have historically been against anything that changes the content of my page, including anonymizing sites that insert banner ads.
I find your reply argumentative, ludicrous, misdirecting, and ill-thought. Your presumption that I am against this because it is Microsoft is incorrect. I was against this idea when it was presented by NBCi, or whoever it was that had those lame commercials 6 months ago with the same idea. The concept of a caching proxy such as Squid altering content is also rather bizarre. As for software that filters ads, I do not and have never used them, and have publicly spoken out against them for the same reasons.
If MS had made this a "feature" that could be turned *on* by the web page using a header tag, I would not be against it. However, they are claiming our content their property by default, in essence; they can modify the context of the content by modifying the definitions of the words within to suit themselves. I find this offensive and somewhat frightening, considering their definitions for the GPL, innovation, and competition.
Most web browsers are designed to present the *content* of my website unaltered-- you are mixing up presentation and content.
Actually, I am not, or at least not everywhere ;-)
You see, images are content. Lynx doesn't show images, and in any case does not show them as intended (embedded in the page). I suppose you are against users of lynx accessing your page?
Babelfish modifies every word of your content, by translating it into another language.
Junkbuster modifies the content by removing a part of it (the ads).
Nanny software prevents users from seing the page, thus hiding the content.
A caching proxy DOES alter content. If it is configured to hold a cached page for a minimum period of time, it prevents the access to the updated page. It is misrepresenting your content, is it not?
And you have no way to prevent it except block proxies somehow (or use a no-cache pragma and hope it sticks)
Compared to any of these, SmartTags, while still a modification of content, is a LESSER modification, in that it does not change what you write, but "comment" on it.
If you are against that lesser modification of content, you should also be against the larger modifications of content, should you not?
I fail to see how my argument is ludicrous, misdirecting and ill-thought, although I can see it as argumentative, and intentionally so.
They are not claiming your content is their property, or at least not more than a dozen others already are.
I would prefer if this feature was configurable by the user, and not by the webmaster, but then again I am way more of a user than a webmaster, so my bias is probably showing.
Then again so is yours, and I managed to show it in a more coherent way, IMHO ;-)
Like others have commented, the implementation is bad, not the idea. I think the idea could very quickly lead to a GPL-style reversal. Think of this:
The user installable Google toolbar for IE already has a feature to highlight the search terms on any web page. It should be relatively simple to alter that so it makes any of the highlighted terms links to searches of those terms. The Google Search Browser Button (for Netscape and IE) -almost- has this functionality already. If it jumped straight to the "I'm feeling lucky" result it would be equivalent to the MS smart tags from a user perspective. MS would have no control over what the resultant link is -- if the smart tag service provider is Google or another search engine which doesn't accept money to place links higher in the search results.
A lot of people inherently mistrust big organizations, be they corporations or governments. I wouldn't be at all surprised if some mozilla hacker put together a smart tag app that automatically linked/searched for the sites which trash rather than praise the selected keyword. Major OEMs like IBM, Compaq or Dell could also substitute their own smart tag modifications (for quite some time Dell shipped with IE And Netscape both installed, so the precedence is there).
All this is not to say "don't worry about it". With the peer review and commentary which is happening now, Microsoft may not necessarily get the results they want. P-)
Authors who think they have control over how their words/presentation/content are presented have gotten spoiled by the distorted browser market.
Flash back to 199x, where Microsoft and Netscape are doing deadly battle. Neither one is really charging browser users money. So the people MS/NS are really trying to get are the "content providers" of the world--- they're the ones that pay for the fancy web servers that IE/NN are loss leaders for. And as a result, the features and overall balance of power in NN/IE are slanted towards the server side.
Let's take cookies for an example. NN/IE do allow cookie blocking, because users were seriously whining, talking to Congress, etc. But turning on "ask before accepting cookies" is incredibly obnoxious, and seems designed to discourage users from using this. Every other significant browser that has cookies has a better implementation. Heck, Lynx has "Yes/No/Always/Never" which lets you say "never" to doubleclick and "always" to slashdot.
So why are Opera and Omniweb better behaved, from a user's point of view? Because the user is their customer. They'll give me just about any feature to make my life better, because that's their primary source of sales. They're not trying to sell OperaServer 2001 or whatever.
And then there's Mozilla. Even if it means forking, Mozilla is gonna get more and more user-slanted functionality---after all, I have control over all the other software on my Linux box. Who are you to tell me I can't implement SmartTags, ThirdVoice, or any other glossing system? It's my machine, dammit.
I think it's important to note that naïve users aren't going to understand that these "Smart Tags" are being generated by their browser: they certainly appear to be part of the page.
That's why the complaint that Smart Tags modify authors' web pages without permission is valid.
I did some volunteer tutoring at the Minneapolis Public Library, helping people who basically new nothing about computers get going on the Web. They were usually thrown by the banner ads that looked like Windows message boxes, even when they had seen plenty of banner ads before encountering one like that. I'm sure they would think Smart Tags were part of the page, even if they "act a little different".
I don't think Microsoft is particularly interested in clearing up these users' misconceptions, either, since the objective is to drive more traffic to sites they sponsor. More legitimate appearance == more hits.
I may be mistaken, but my impression is that their "off by default" decision only came after they began to catch a lot of flak about this proposed "feature".
New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.
Keep up with the latest Advogato features by reading the Advogato status blog.
If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!