What's left for RSS to disrupt? Plenty

What's left for RSS to disrupt? Plenty

Summary: Updated 5/20/05: Steve Gillmor is back on "the air" again.   Under the name the "Gillmor Daily," Gillmor's new regular gig featured Dave Winer (credited with the birth or rebirth of the XML-RPC, RSS, podcasting, and OPML) for the first show and based on what is said near the show's end, Winer will be making regular appearances.

TOPICS: Tech Industry

Updated 5/20/05: Steve Gillmor is back on "the air" again.   Under the name the "Gillmor Daily," Gillmor's new regular gig featured Dave Winer (credited with the birth or rebirth of the XML-RPC, RSS, podcasting, and OPML) for the first show and based on what is said near the show's end, Winer will be making regular appearances.   The exchange that most caught my attention came at 15:05 minutes into the show:

Winer: If you're expecting more from RSS and you're expecting more from the Web, I think you may be disappointed.   I don't know that it has much more to offer.  

Gillmor: I'm not sure I agree with that.   

Winer: I still like you though.

I'm in Gillmor's camp.

[Correction: Via e-mail, Winer explained that I incorrectly took his statement out of the context that immediately preceded it.  That conversation involved a discussion about generating revenue off of feeds and Web sites.  Contrary to the context that I put his statement into, in the e-mail, Winer said  "I expect a lot more from the RSS and the Web."  Most of what follows goes into some of what I think we could see from RSS.  But where a different context to Winer's statement is implied, I've attempted to correct it with strikethrough.  I also changed the headline which originally read "No more tricks for RSS? That depends"].

The future of the Web is a discussion unto itself. For example, how will the user experience change when and if everyone has gigabits of both uplink and downlink bandwidth as opposed to kilobits, which is what most everyone has today.  [Added: Over time, bandwidth has traditionally affected design standards (eg: acceptable page weights in bytes).  Whether or not additional bandwidth might lead to innovative changes in online revenue fundamentals still remains to be seen.]  So, I'll leave the Web discussion for another day.  But I don't think RSS has hit the end of its rope.  Not by a long shot.  [Added: Beyond the discussion of revenue models, RSS has some other, highly disruptive potential.]   It's just a question of whether some people have the guts to break old habits.

One of those old habits is Internet's e-mail protocol known as the Simple Mail Transport Protocol (SMTP).  People keep saying there's nothing that comes close to being a silver bullet to the spam problem.  Not true.  RSS would be like an intercontinental ballistic missle fired right into the lion's den.  If tomorrow, we replaced SMTP with RSS, spam (at least e-mail spam) would probably disappear for good.   There are two great reasons this is true. 

The first of these is that RSS is 100 percent opt-in.  You won't get an RSS feed in your RSS reader unless you ask for it.   This means that we get to decide who can send stuff into our inboxes. This is much better than the system today where Congress and the FTC apparently feel they have the right to decide that (while the big money marketing lobbies magically get more of the lawmakers' ears than you or me).  

Reason #2: Whereas SMTP is a store-and-forward protocol, RSS is a store-and-get-retrieved protocol.   With SMTP, by the time the spam is in the mail (most of the time with a bogus return address), the spammer has taken the additional step of closing up shop and moving.  The way spammers do this reminds me a bit of those oriental rug shops that appear to go out of business within days of opening.   One day, someone will explain to me why this is a sensible way to sell rugs (I'll bet it has to do with a loophole in an import/export law).  I digress.   With RSS (the "Retrieving Stops Spam" protocol), if the spammers are serious about actually getting the spam into our inboxes, then, not only can't they give us bogus address information, but they also have the burden of storing the spam on their systems and, even more importantly, have to set up an XML feed that points to where the spam is stored.  Today, SMTP allows spammers to go out of their way to cover their tracks.  Tomorrow, if we replaced SMTP with RSS, spammers would have to lay tracks that lead straight to their front door.  

To me -- a person that has tirelessly campaigned for more cooperation among the industry players that are actually empowered to curtail spam -- when I think of the potential of RSS, I'm not sure today's usage of it even scratches the surface.   

To a lesser extent, I think the idea of multiple enclosure support will open up a whole new world for RSS as well. Enclosure support is what facilitates the distribution of rich media (and data) via RSS.  Today, where enclosures are supported and in use (for example, with podcasting), the number of enclosures that can be bundled with one feed item is one.   Technically speaking, it probably could be two or more.  It's just that the software on either end of an RSS feed isn't really ready for that world.  If I had a suggestion for the newly established Podcasting Specification Working Group (shown to me during IDG World Expo's recent Syndicate Conference by Hearst Interactive Media vice president Michael Dunn), it would be to get the multiple-enclosure idea into a proof-of-concept stage.   By now, some people are asking "why would you want to do that?"  Here's why.

RSS is largely about distribution of ideas.  I say "ideas" as opposed to content in deference to forward thinkers like Doc Searls (need a really good laugh? read this) and David Sifry, both of whom dislike the phrase "content consumer."  As an idea generator, I currrently transmit my ideas as text and/or audio.  And RSS supports both of these quite nicely (text naturally, and audio as an enclosure).  But, there are plenty of applications where the idea may require medium types that are different from (or additional to) text and audio.  Video comes to mind.  But, pursuant to the spam issue that I just discussed, what about e-mail with enclosures that can consist of just about any special file type? For example, PDF and/or DOC documents (perhaps secured ones),  spreadsheets, JPGs, GIFs, vCards (partially handled now within the RSS specification), etc.  

Actually, this is where RSS gets dangerously close to --and perhaps encroaches on -- collaboration, which is mostly dealt with by today's solutions in a proprietary fashion.  In an interview earlier this year, Userland CEO Scott young taught me why it makes much more sense to use standards-driven subscription-based infrastructures (aka: RSS-driven infrastructures) as opposed to proprietary ones.  Looking back at Ray Ozzie's blog for example, I think it's interesting how, after pioneering corporate blogging policies in 2002, his blog eventually disappeared into the ether.  Ozzie, as most people know, runs Groove Networks, which was recently acquired by Microsoft. 

Groove makes one of those collaboration tools that's largely architected around the idea of projects, workspaces (which include documents), participants, and security.  The idea of RSS-driven collaboration (distribution of ideas, documents, etc.)  is completely disruptive to the Groove-like crop of collaboration tools.  For example, workspaces and contributions to them may often involve documents of different types and this is precisely the sort of activity that RSS with multiple enclosure support could serve. Ozzie's blog, as it turns out, could have been proof of concept that Groove's approach was already behind the times.  There was Ray Ozzie, sharing ideas and collaborating with others and what was his tool of choice?  Well, it certainly wasn't Groove.   Survival in the software business often means being careful about eating another dog's dog food before you eat your own.  Even if the other dog's food tastes better (and the other dog is willing to share).

So, again, if you ask me, when it comes to RSS, you ain't seen nothing yet there's huge potential for it to disrupt certain status quo technologies and approaches to business problems.  Perhaps we never Whether it will (or not, which would be unfortunate) may depend on who has the guts to change. 

Finally, although the show is currently available through Gillmor's blog on ZDNet, longer term The Gillmor Daily will be a part of Adam Curry's stable of offerings via his new Podshow venture.  Gillmor discusses the arrangements during this first episode.  We wish Steve luck as he embarks on his new journey.

Topic: Tech Industry

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • Why is it that Dave is always complaining about being mis-quoted?

    Because he spews out garbage without thinking and then later complains about being mis-quoted when he realizes what an ass he sounds like...
    • ????

      Puzzling - why would you waste your time reading something that you feel provides no use?
  • What is RSS?

    Usefull to know to discuss this
    • Spoon fed


      I used the "search" field to dig it up.
      • Bad Google, no URL!

        Good articles inform the reader about the issue; they don't force you to google the subject to understand the subject.

        - Journalism 101
        • On the other hand...

          ...how much hand-holding does an author need to do, anyways? Besides, this is not about what RSS is, but more specifically about the extent of its reach, with the understanding that a reader of this article knows something about RSS to begin with. It's not unreasonable to assume a certain level of knowledge and competency beforehand on the part of the reader, particularly from a visitor to a site like zdnet.
  • RSS

    An education of sorts. RSS developed into a more powerful tool makes sense to me. I would rather opt-in on email than continue to receive the junk that arrives daily.
  • I disagree with two points in the article

    Author states: "The first of these is that RSS is 100 percent opt-in. You won?t get an RSS feed in your RSS reader unless you ask for it.... This is much better than the system today where Congress and the FTC apparently feel they have the right to decide that (while the big money marketing lobbies magically get more of the lawmakers? ears than you or me)."

    I'm in disagreement of two points. The first one being the obvious political one, and I'll only comment on it quickly to keep things from going out of control...if they have a better system in mind, then they damn well have the right to deserve that. Who gives a f**k if they work in the government? Second point, I disagree in RSS email being better than regular email overall. Companies will be shot to death from this, since if they need feedback or have a troubleshooting bug that needs to be reported by a user, then both ends need to add manually each other's email addresses to do that. I thought email was supposed to be automatic and computer-oriented, not a manual labor hog like RSS email would suggest. The only people who will benefit from RSS email will be the few people who like living in houses protected by electric fences and barbed wire in that only those people who know who to invite will benefit. Therefore, there will be an RSS email trend to deal with in the near future.
    • re: your two points

      1. I'm not sure I understand the gov't comment. the feds are passing laws that affect the way we use technology, and are giving hall passes to there more supportive constituents.

      2. Imagine if, the first time you received an e-mail from a particular source, it validated the feed if the feeds checked out, it prompted you to subscribe. So, you'd only subscribe once. And, if at any time, you get abused, you just unsubscribe.
      • Grammar Check...

        [b]...are giving hall passes to [u]there[/u] more supportive constituents.[/b]

        That should read:

        "[b]...are giving hall passes to [u]their[/u] more supportive constituents.[/b]

        THERE: not here.
        THEY'RE: Contraction of "They Are"
        THEIR: Possessed by "they" - whoever they may happen to be.
      • Unsubscribe?

        As you said yourself, currently spammers move around a lot. Trust me, they will find a way to exploit parts of that new infrastructure. What if that exploitation is similar to open relays and they use valid servers, or find a way to spoof oft-trusted servers, like AOL? You can't just unsubscribe to AOL, especially if you have many friends on it.

        Are we trading one set of holes for another, and at great expense and bother? Whereas working on improving SMTP seems like a smarter approach; if some SMTP solution turns out to be worse we wouldn't be married to the new system like we would be after investing so much in RSS.
  • "Disrupt" is right - RSS is unsupportable from the server side

    We need to have a serious discussion about the dark side of RSS. I have commented in previous Talkbacks about the "pull" nature of RSS. Contrary to popular opinion, RSS is [u]not[/u] push technology. Each client of an RSS feed has to poll the server periodically to see if there's anything new. That polling operation, times the number of subscribers, times the frequency with which they poll (the last not controllable by the server), adds up to a tremendous server-side load on your hypothetical mail system.

    Store-and-forward might look pretty nice by comparison, spam and all.

    I *think* there's a hole in the logic that suggests RSS-based mail would be spam-free by being totally opt-in. But I never seem to get far enough into a real analysis of how it would have to work and where the loophole might be, because I get hung up on the server load issue. If you ask me, the whole concept is a non-starter.
    • RSS is scalable, according to Bray


      May I direct your attention to exhibit A:


      I trust Bray on this.

    • There's NO such thing as foolproof...

      [b]I *think* there's a hole in the logic that suggests RSS-based mail would be spam-free by being totally opt-in.[/b]

      Exactly. How long do you think it would take Joe Spammer to figure out a way to pollute RSS feeds with spam. As the saying goes:

      There's NO such thing as "foolproof" because fools are so ingenious.

      Ok.. Say I'm Joe Spammer. I want to put spam in your RSS inbox. So I go to any of number of free or cheap web hosting sites out there. I stick my content up on the site, I taint the feed and send it on it's merry way.

      It gets taken down three days later? So what. By that time I've gotten my message out to hundreds of thousands. I've moved onto the next free/cheap server and the next round of spam.

      And I'm NOT even out a penny. Why? Cause I used someone else's credit card info to set up the account.

      Not bad for 5 mins worth of thought.
  • Nothing is absolutely secure from abuse

    Without doing a complete analysis, I'm only shooting from the hip. However, I believe that nothing is absolutely secure from abuse. I can imagine some bright but less than forthright individuals taking over various servers with valid RSS info and substituting their own. Conversely, what about DDOSing RSS sites through major overloading. This may well prevent a lot of the common nefarious schemes of today, but I suspect that it will open up a new batch for tomorrow when it becomes prevalent.
  • Distressingly narrow view

    I find your view of the email system rather limeted. Sure RSS is a fine replacement if all you do is recieve newsletters or the weekly sale listing from your local computer store, however email has many more uses. Not long ago I was delited to recieve one of those "are you the person who was at..." messages from an old friend from the service whome I had been trying to locate for years. Your system would have made that very difficult for both parties, and probably not worth the effort. At work I am contacted by customers needing information about the charactistics of our product, sometimes years after the project we worked together on. On occasion someone I have worked with will give my address to a coworker, customer or to someone who had a question they couldn't answer. Some of thes3e "third party" questions have been very interesting and fun to sort out.
    There is much more to the world of e-mail than the distribution of newsletters and I hope RSS lives alongside of e-mail and doesn't replace it.
    • You're mistaken

      99.99 percent of the complexity of moving from store-and-forward to post-and-poll could be easily dealt with. This is not just for newsletters. This would be for all e-mail and I don't think you can point to a particular technical hitch that couldn't be easily managed.
      • An Example?

        Let's say that I'm a small guy that runs a listserv for a wireless users group in San Francisco with 2000 subscribers; I have Business Class DSL and a small server, say a 3.0GHz P4 with 2GB RAM.

        Currently this is enough to handle my listserv's email messages, my own family email, and some small websites. The traffic and processing is currently well within the limits of my hardware and internet connection - but won't moving to RSS require that I have a much faster connection and server to handle 2000 users "checking" their email (with maybe 200 users trying all at once)?

        Consider: Instead of a load that sees the sending of 2000 emails 100 times a day (something almost any server can handle) it turns into me being an email server for 2000 people, say 1500 active users, checking for mail every 1-5 minutes (which is how often most people I know check their email)! That seems like an insane amount of traffic for a small list serv to handle ? 10 users a second, all day.

        And that's just for the list serv, it doesn't include the 50+ people I'd have to check and the 50 or so my wife would be checking; all on top of the 100+ each of my two kids have in their email contact lists. That's 300 servers we?d all be checking all day, and another server a second we?d have to check on top of it if we check every 5 minutes.

        Now imagine that my kids are downloading an MP3 from iTunes. And someone else is checking out my wedding photos on one of my in-house webpages.

        Suddenly, the setup that's worked flawlessly for years is now completely overloaded with what should have been simple email traffic.

        Wouldn't this make it harder for the little guy to compete, even on a hobby level? Or does the little guy not count?

        So, how about an example of how this is easily overcome without forcing the small guy to spend a lot more money.
      • Doesn't AOL already have the same results?

        "Post and Poll" seems very similar to what AOL already offers with the existing SMTP system. You can set up filters that will sort through your mail box and automatically dump junk mail. You can delete the rest without having to open it (thus exposing your machine to malware).

        It seems to me that if I had to go through my RSS list and constantly add/delete names, it would be more of a hassle than it is worth.

        Sorry - but there are other options out there that make more sense. IE: Having mail servers scan documents and automatically dump documents that are duplicated more than say 10 times in an hour.
  • White-listing?

    If opt-in is the reason RSS is so good for spam, wouldn't white-listing achieve the same result, but with less infrastructure changes and needless application development?

    Not that I'm a fan of white-listing either, but why complicate the issue so much if all you need is opt-in?