Action Cards change everything; Greasemonkey is a Firefox only hack.
Paul Madsen blogged about his rejection of how the recently announced Action Cards—specifically the ChoixVert card—changes anything. He further posits that Greasemonkey can do everything a selector-based Action Card can do for browser augmentation.
Since I wrote the white paper he refers to, and the ruleset for the ChoixVert card, it is appropriate that I respond.
My position is that greasemonkey is a Firefox specific hack that is a fun toy and is no more than a harbinger to the real thing. Selector-based Action Cards are the real thing and do—in fact as claimed—change everything.
Paul states that “client-side personalization of search results is undeniably cool” but that this coolness was changed a while ago with greasemonkey. If you have read the whitepaper, you will be able to read between the lines that I anticipated this reaction from those who were too occupied to dig and understand the facts. Sure enough, here we have a blog post making ignorant claims.
Paul then jumps to the ChoixVert whitepaper and lists why I say the selector-based web augmentation is way better than any others. When I referred to “many scripting languages and web augmentation tools” I didn’t spell out greasemonkey, but that was one of the specific browser augmentation tools I was talking about.
In the whitepaper, I list the following reasons why selector-based augmentation is superior to client-side scripting:
- KNS is selector driven
- KNS is accessible. (human readable)
- Secure glue to accessible data to any service or application.
In his blog post, Paul addresses two of the points found in the white paper. The case he makes is wobbly and uninformed, but here are his points.
Paul’s Claim
Being selector driven is at a disadvantage, as the user cannot customize the Information Card.
My Response
Cardholder access to the ruleset would reflect poor design. It isn’t done for a reason. Access by anyone to any card’s ruleset is fraught with huge danger. Any hacker could completely destroy the integrity of your card for whatever reason.
The way Kynetx KNS is set up, any person can generate their own card and write their own rule set. The ruleset and card then become signed to each other. You can’t have a card from one ruleset and be running another (spoofing or phishing). I much prefer this approach than letting anyone edit the ruleset.
Of course this is one of the reasons greasemonkey is so dangerous, anyone can hack a greasemonkey script and the script user would be none the wiser. This is why greasemonkey is a hack. Smart people came up with a quick and dirty way to augment the browser. Very cool. Very hackable. If hacked, you’re screwed, there is no viable remedy but to start over.
Here is the bottom line difference. With Action Cards, smart people architected a well thought out, scalable, experienced, sustainable, manageable, secure method of augmenting the browser and the operating system. Very, very cool. Very, very tough to hack. If it is hacked, there is a manageable remedy.
Paul’s Claim
Paul presents the cliché “His mother is reassured that rulesets are readable.”
My Response
While I discourage the use of clichés, it is kind of funny; but irrelevant. Obviously I wasn’t referring to Paul’s mother or mine when I stated that KNS is accessible because it is human readable. What I meant by that is that almost anyone (not your mother) could look at a ruleset and get an idea of what it is doing. In other words, creating rulesets is not a prohibitive activity. I could have been clearer on the issue; I don’t think I needed to explicitly exclude his mother.
My point was, a person capable of writing a greasemonkey script could write an action card ruleset.
Paul doesn’t address the third point. He does so for good reason. Greasmonkey has no capability, or possibility of being the data glue for any service or application. It is a hack that can use web-based data for browser augmentation on Firefox. Cool, but not earthshaking or does it change much.
Let me add a few points to each claim that I made in the whitepaper.
To refresh, I stated that web augmentation based on KNS is in a class of its own. And that other web augmentation tools (specifically Greasemonkey) don’t hold a candle to the KNS implementation. Here is more on each of the three reasons stated.
Selector Driven KNS
KNS is selector driven—obviously, greasemonkey is not. Here are the reasons why a selector-driven augmentation tool makes greasemonkey look like child’s play:
- Security—there is no question that the Azigo/Kynetx architecture leaves a bigger security opening than if the information card was a traditional personal or managed card. This is because Kynetx uses an Azigo provided API to the selector to query which card to assign a specific ruleset. This is why it is called an “Action Card.” However, every information card—even an action card—is digitally signed. Thus an action card can’t be spoofed or phished. If a mal action card is discovered, it can be banned or stopped at the server so it will not work anymore, anywhere.
- Cross-browser –greasemonkey is a Firefox only thing. For all of us avid Firefox users that works. For much of the planet, it doesn’t. You can get a selector for almost any operating system/browser combination. Soon, you will be able to for any combination. In the future I expect selector access to be built in to all browsers.
- Cross-platform—binding data to a browser experience is one thing. Information cards can be used for authentication and data relationship management for applications and cloud-based applications. No greasemonkey is this space.
- Cloud-based—have the ruleset exist and run in the cloud brings lots of advantages. It makes it much more manageable and scalable. It also provides a much greater level of control. For example, if there is a bug in ruleset, you simply change it in the cloud and every cardholder automatically benefits from the fix. Nothing needs to change on the client, no reinstallation, just clear sailing.
- Auditing and Analytics—KNS is way cooler than greasemonkey as it lets the ruleset developer access analytics and auditing for the ruleset. As any good SEO person would know, good records of what is happening isn’t an option if you want to be on top your game. Greasemonkey has no such notion. This point alone makes greasemonkey look like the toy that it is.
Accessible Technology
Simple User Interface—while it isn’t likely that my mother will write a rule, she—or my father—can install an information card. I doubt that finding and installing a greasemonkey script is anything she will ever do.
Data Glue
This is the one point I made the Paul didn’t bring up. I can only speculate as to why. But let me say this:
Greasemonkey can’t be browser and platform independent glue for any application or service. KNS does this so elegantly and specifically. We haven’t seen the tip of the iceberg yet.
Summary
Web augmentation is an incredible phenomenon that we are just beginning to understand and use. There is a spectrum of tools available to accomplish various levels of augmentation. I only talk about two of those here. Greasemonkey and Action Cards.
I stand by my position that Action Card web augmentation changes everything. And that greasemonkey—at its most lofty view—is a mere harbinger of the real thing. Greasemonkey lets you do basic web augmentation with lots of potential problems and drawbacks.
Action Cards—the combination of the selector-based information card, KNS, and cloud-based data is elegant, well thought out, and well architected capable of making long lasting significant changes to the Internet.
Don’t get me wrong about greasemonkey though. I love greasemonkey. I use it every day. But I love it for what it is. I don’t expect it to be something other than that. To pretend that it goes beyond its design is a disservice and uninformed.
Greasemonkey is a hack for changing Firefox browsing and chrome. An easy fun hack. But a hack nonetheless.
Look for Action Cards to change everything and greasemonkey to be a fun toy script for hackers.