Go to previous post:
Red fish, blue fish.

Go to Electrolite's front page.

Go to next post:

Our Admirable Sponsors

January 13, 2004

Technical blip. Okay, so why would modifying my RSS templates so that they syndicate excerpts only, rather than full posts, cause my RSS feeds to go completely blank? Let’s see what happens when we add a new post.

UPDATE: Okay, so once you rebuild and add a new post, the feeds regenerate themselves. Learn a new thing every day. [02:25 PM]

Welcome to Electrolite's comments section.
Hard-Hitting Moderator: Teresa Nielsen Hayden.

Comments on Technical blip.:

Rick Heller ::: (view all by) ::: January 13, 2004, 04:56 PM:

Until yesterday, I didn't have an opinion about full text versus excerpts. But having set up an aggregator which includes your blog:


I'm glad to have a partial available. The full versions clog up aggregators, and defeat the purpose of being able to monitor what's going on at a glance.

In fact, I've deleted most of the full feeds I was subscribed to.

Bill Humphries ::: (view all by) ::: January 13, 2004, 06:09 PM:

Just so you know, you're not limited to choosing either full feeds or excerpts. You can have multiple syndication templates in Moveable Type.

And to gently disagree with Rick Heller above, I've decided, on reflection, that I prefer full feeds. I used to feel the opposite way, but I don't think full feeds 'clog' my aggregator. I'm using Net News Wire, so my feed view is like a good newsreader or mail client.

Now, if you're using Radio as an aggregator, I can see where full feeds will clobber that page (and broken markup will hurt it too, like a Live Journal friends' page after someone posts a poorly designed quiz result.)

However, in my opinion, Radio's a better publishing tool than aggregator.

Seth Morris ::: (view all by) ::: January 13, 2004, 06:54 PM:

I've been reading the RSS feed for a few weeks. I run an aggregator which pulls articles to my PDA and (offline) laptop. Partial feeds are basically useless to me. I'm in the process of cancelling subs to partial feeds, although I do click a few now and then.

I just don't understand the value of a partial feed: how is getting less data a good thing?

Obviously, I'm missing a common use case where having the full article is bad. I'm inclined to blame the tools, not the data format, but it's certainly easier to restrict the data that to rewrite the tools.

Oh well, been nice reading you while it lasted! Keep up the good commentary.

Douglas Johnson ::: (view all by) ::: January 13, 2004, 07:27 PM:

I much prefer full posts--I was very pleased when you went to full posts in your feeds some months ago. I read everything you post, so teasers are just an annoyance for me.

Regardless, I'll read your blog anyway.

Patrick Nielsen Hayden ::: (view all by) ::: January 13, 2004, 07:42 PM:

I'm afraid this played a non-trivial role in my decision to stop syndicating a full feed.

Douglas Johnson ::: (view all by) ::: January 13, 2004, 07:56 PM:

Oh. Point taken.

I can live with taking an extra step to read your posts.

Thanks for taking the time to explain--I'll never crab about truncated posts again.

Bill Humphries ::: (view all by) ::: January 13, 2004, 09:02 PM:

Net News Wire did the 'show differences' thing when you went to excerpts, but that was a side effect of changing the RSS feed on the previously existing posts.

Mithras ::: (view all by) ::: January 13, 2004, 09:49 PM:

I much prefer full-text feeds, because there are just too many blogs to visit each one. I read 258 feeds. I just end up ignoring the posts unless the excerpt is really, really interesting.

Damien Neil ::: (view all by) ::: January 14, 2004, 02:14 AM:

I also use NetNewsWire Lite. Full-text feeds are much more pleasurable to read than partial-text ones.

I'm not quite certain what the inessential.com link is supposed to be telling me--are you dropping the full feed because you don't want people to be able to view differences when you change a post?

Suw ::: (view all by) ::: January 14, 2004, 03:58 AM:

Aw... but I liked the full feed! It's not that i'm overly consumed by dozens of rss feeds, or that i'm lazy, but it's so much easier when everything pops up in my aggregator and is all there and ready to rock and roll.

Maybe you could run two feeds and keep everyone happy?

Kevin Marks ::: (view all by) ::: January 15, 2004, 05:13 AM:

please bring it back...
Is showing your edits giving away professional secrets of the editors?

Patrick Nielsen Hayden ::: (view all by) ::: January 15, 2004, 08:20 AM:

It has nothing to do with being an editor, and everything to do with being a writer. So let's knock it off with the "secrets of the editors" stuff, thanks very much.

Brent Simmons's addition of a revision-tracking feature to NetNewsWire means that any minor tweak I make to a post after it goes up will now be colorfully highlighted in for all his users to see, so long as I syndicate a full-text feed. I gather that there's a well-known weblogger who some people feel is in the habit of rewriting his older posts to make it look as if he did better in certain arguments than he actually did at the time. That's very aggravating, but that guy isn't me, and I find that the idea of having my little cosmetic fixes put under a giant spotlight has a radical effect on whether I find this particular hobby fun. Yes, of course, "diff" has existed for decades, and yes, of course, someone may come up with a similar plug-in for regular old browsers. You know something? If they do, and if it starts being widely used, I'll probably stop writing this altogether.

You can call it vanity if you like, or you can lecture me about what information wants to be. Suit yourself. But here's a clue: Technology doesn't always make life better. It's value-neutral. Some innovations have the effect of stimulating people to make creative works and give them away. And some innovations have the effect of stimulating creative people to clam up and go do something else. I enjoy writing a public weblog, and I don't mind if my small revisions are occasionally noticed here and there. But having a giant spotlight trained on them turns the whole process into a very self-conscious one. I don't need that. I have other things I could easily do with the time that goes into this.

Jeremy Leader ::: (view all by) ::: January 15, 2004, 07:54 PM:

Patrick, as a non-rss user (and hence, almost completely ignorant about the tastes and behavior of rss users, so take this for what it's worth), I suspect that one of two things will happen:

1. Rates of update to blogs will go way down; either habitual updaters will stop blogging, as you threaten (and yes, to some of us, "no more new Electrolite articles for you" is a threat (and I hope you recognize that that's a compliment, albeit backhanded and parenthetical)), or they'll keep blogging, but update less, because it doesn't do much good to make corrections that leave the error visible.


2. People will mostly ignore the revision highlighting. Can it be turned off? If not, people will either stop using NetNewsWire, or train their brains to mostly ignore revision highlighting, just as they've trained their brains to ignore anything banner-ad-shaped. After all, as you imply, most revisions really aren't that interesting, and having them highlighted usually distracts from the point of the text.

Or (ok, make that 3 things):

3. Someone will come up with a way of tagging revisions to tell NetNewsWire "don't highlight this change, it would just be distracting" (though I'm not sure how I'd design such a feature). Of course, then someone will want a way of telling NetNewsWire "highlight revisions, even ones you're not supposed to".

I've come to the conclusion that all communications technology is a negotiation, at least when mediated by software. For example, think about things like font specifications in HTML. The author wants the text to look a certain way, but the reader has their own ideas about how they want to look at it.

Actually, I just thought of a form of revision highlighting that might be even more distasteful to authors; imagine if an editor's changes were highlighted, and attributed to the editor. So every time an editor corrected something, it wouldn't just be a private embarassment to the author; it would be like publishing a school assignment covered with the teacher's corrections in red ink. Of course, that assumes that the author isn't a nutcase who thinks all editorial changes degrade hir work.

Dave Slusher ::: (view all by) ::: January 15, 2004, 11:52 PM:


I've weighed in multiple times that I think a full feed is the way to go. Multiple feeds are even better if you feel strongly that you must provide a truncated one but if you are to only have one, full is the way I'd prefer to see it go. A full feed can always be excerpted, a clipped feed can never be expanded.

A lot of the argument for clipping comes from the presumption of how the data feed is being used, such as "The full versions clog up aggregators, and defeat the purpose of being able to monitor what's going on at a glance." The whole point of RSS is that you don't know what's happening at the other end of the pipe. A lot of us use aggregators that are not intended to monitor "at a glance" but are an alternate means to read the full articles. The feed could be getting indexed, or put on a PDA, or whatever. If it's truncated it limits the utility for many of those uses.

It's your party, cry if you want to, but I just don't see the existence of this feature in NetNewsWire as a decision maker for bloggers. Who cares? It's the tail wagging the dog, dumbing yourself down because of the feature set of one client out of the gazillion clients out there. It's like writing bad HTML code to fit the rendering quirks of browsers - people do it because they think they have to at the time then the client changes the behavior and everyone is stuck doing crazy shit for no reason. You know, if you have to make an edit at the beginning of a post, it will still show up in NNW anyway. Fight the power, don't clip your feed because some biscuithead adds in a highly annoying feature most people will turn off anyway (or if they can't turn it off they'll quit using NNW.)

From my experience aggregating with FeedOnFeeds, I can tell you that blogs that make me do extra steps to even figure out what the goddamn post is about are blogs I stop reading. For me, the aggregator is not a reminder to go check out the web page, it is the way I prefer to read the posts. I have limited time for my aggregation reading, so if I can't get read it in one step directly from the FoF page the odds of coming back to it decrease rapidly. If I only have a title and a few words to go on, the odds approach zero.

Nicholas Liu ::: (view all by) ::: January 17, 2004, 09:26 AM:

I don't use an aggregator, but it seems unlikely to me that most people would turn on this feature except now and then out of curiousity. Why would they want to see every edit you make? That'd just be distracting. Anyway, I don't think anyone (except people already giving you a hostile reading, in which case it's hopeless anyway) is going to think any less of you for editing old posts, and I don't see why they should.

Mitch Wagner ::: (view all by) ::: January 17, 2004, 07:47 PM:

Patrick - You know that I'm a fan of running complete feeds instead of excerpts, but I'm a bigger fan of Electrolite, so whatever you do, I'll keep reading here.

You can't get rid of me that easy.

The NetNewsWire feature doesn't bother me -- I do occasionally make revisions to posts, but I don't think they're anything that anyone will be interested in.

I just started a section in my own blog for drafts and notes -- preliminary versions of things that I plan to make full-scale posts out of.

There's a controversy in the exciting world of online journalism: if you make a correction in an article, should you make a note that material was corrected?

Kevin Marks ::: (view all by) ::: January 19, 2004, 04:07 AM:

Sorry about coming across as snarky - I did like the full feed, and I do miss it.

NNW's feature has to be explicitly turned on, and you shouldn't let its existence stress you any more than diff does.

Cory's habit of showing substantial corrections in BoingBoing using strikethrough is a good one.

Patrick Nielsen Hayden ::: (view all by) ::: January 19, 2004, 09:12 AM:

There's a big difference between a command-line tool with a steep learning curve such as diff, and an easy-to-use feature in a popular graphical-interface reader. As the world by and large understands.

I personally try to highlight very clearly when I've substantially changed my mind, or discovered I was wrong. That's not what this is about.

Jeremy Leader ::: (view all by) ::: January 19, 2004, 02:43 PM:

By the way, over the weekend I was required to edit an MS Word document I'd written earlier, with change tracking turned on. Man, is that irritating. Patrick, I now understand completely where you're coming from!

However, I still think blog corrections will become very rare (I don't know how common they are now), or people using NetNewsWire are going to turn off the diff feature 99% of the time. I think change highlighting is going to be almost as irritating to the reader as to the writer.

Another data point: some wikis use revision control, and provide a way to see the history of changes to a page. For example, kwiki lets you choose whether you want to see a list of past versions at the bottom of the page, and further whether you want to see the most recent changes below that, with options to examine earlier changes as well. However, by default it shows none of that, and even when you turn on all the bells and whistles, it first shows the current version of the page, unmarked, with the changes below. So you always get to read the latest version first, and the change highlighting is always secondary. When you're creating text communally, as is the WikiWay, seeing the history of changes helps prevent see-sawing changes, or going around in circles.

Mitch Wagner ::: (view all by) ::: January 19, 2004, 04:01 PM:

Patrick: I'd be grateful if you articulate why you find the NetNewsWire feature objectionable -so objectionable that you're willing to inconvenience a significant portion of your users in order to block that feature.

Please do not be offended by this question - I am not trolling here, I would like to understand your point of view.

Like you, when I make a substantial change to a post, I always make a note of it in my blog -- actually, I do more than that, I put up a new post that simply says, "You know what I said before? Forget it - this is what I think now." I can only think of one time I've done this, but it was pretty big -- I've gone from opposing the war on Iraq, to supporting it, to opposing it again. Once or twice I've also made corrections to factual errors pointed out by readers.

So if someone wants to turn on that NetNewsWire feature for my blog, all they're going to see is a string of corrected typoes and spelling errors and the occasional infelicitous phrase smoothed over. And they won't even see much of that, because my philosophy in blogging is that once something is posted, it's generally a waste of time to fiddle with it; it's better to spend the energy and time on a new post.

I do want to emphasize here that I respect your opinions and I'm not trying to dictate how you run your blog. Yes, I am inconvienced by this decision of yours but it's not like it's some HUGE inconvenience -- I mean, all I have to do is click a thing and wait a second where I did not have to click and wait before. All of the inconveniences in my life should be so trivial.

Patrick Nielsen Hayden ::: (view all by) ::: January 19, 2004, 06:41 PM:

I'm all in favor of revision marks. They're a great way to track change history when multiple people are working on a document -- a writer and her editor, for instance, or collaborators.

However, for writing of this kind, I find the knowledge that tracking my cosmetic revisions has been made trivially easy makes me feel like I'm writing in a store window. Or like people are looking over my shoulder.

It's not enjoyable. I'm reasonably careful to explicitly note when I've changed my mind, or been mistaken, or given out false information. I don't think I need a technological honesty enforcer. And I don't like performing the little esthetic brush-ups as an act of public performance. It takes something that was fun and makes it not fun.

Jeremy Leader ::: (view all by) ::: January 19, 2004, 07:20 PM:

Patrick, I completely understand. What was irritating about Word's revision marks was not that I or my collaborators could see them, but that our audience would be able to see them.

Fortunately, Word gives the author control to selectively turn off the marks, so I can go through and turn off the marks for revisions that are purely cosmetic (for example, the length of the document changed, so in every page's footer, where it automagically says "Page X of Y", the old value of Y is there in red, struck through).

As I said before, though, I doubt most people are going to leave the "show revisions" feature turned on for long. It's just too ugly. It'll probably turn out to be more like writing in a sidewalk cafe, where passers-by can see what you're doing, but few will be sufficiently interested to pay attention. But if you prefer to sit at a table farther from the foot traffic, that's certainly preferable to having you stop blogging.

Mitch Wagner ::: (view all by) ::: January 20, 2004, 11:51 AM:

Patrick - I find it an interesting coincidence that you took the steps you did to reduce transparency in your blog, at the same time as I made a couple of major changes to my blog designed to INCREASE transparency.

To quote from my favorite writer's favorite blog (me, mine): "I started the Drafts and Looks Interesting sections of this blog to increase transparency in the blogging process. The Drafts section will contain the starting elements of items that I intend to write full blog entries about, but don't have time to do so right away. It'll hold URLs, and fragments of sentences and paragraphs that I hope to later include in polished blog entries. I imagine that many of those entries won't ever get finished, which is, for me, part of the fun.

"And the Looks Interesting section is for URLs that look like they might be interesting but which I haven't actually checked out yet. Used to be if I found a really long foreign policy article that I didn't have time to read at the moment, or a long, allegedly humorous streaming media program that I didn't have time to view right away, I'd bookmark the URL on my desktop. But now I figure, hey, why not share those links with my legions of readers as well? ('Legions'? Not 'legions.' 'Squad,' certainly, maybe a 'platoon,' but nowhere near 'legions.')

"I am pleased about this because both of these sidebars move this blog closer to the blogging essence: spontaneous writing, direct from writer to reader, without polishing or editing (or, in my case, proofreading - but I'm working on that)."

I've wanted to be able to create in public since high school. I already considered myself a writer back then. I noticed that nobody wants to watch a writer work. Why should they? Watching a writer work is boring. Well, Julie finds it interesting but (1) she's my wife (2) she thinks it's funny to watch me stare blankly into space with my jaw slack, and then look decisive, focus on my monitor and type furiously and (3) even she seems to only want to watch me write for a minute or two every few months.

But if someone picks up a paintbrush or pencil to draw in public, people will gather around behind him to watch the process. Drawing and painting fascinates people. ("People," at that time in my life, meant "girls," who were the variety of people whose admiration I was most interested in drawing.)

Making music is, of course, inherently a public act. Music without an audience does not exist. When a musician plays his instrument without an audience, we don't even say he's "playing," we say he's "rehearsing" or "practicing."

Later still, pop culture introduced us to the idea of the "rock 'n roll comedian." Comics like Steve Martin and Eddie Murphy could fill an entire sports stadium with an audience, and Murphy, at least, had an entourage and attitude like a rap star. I envied comedians because it seemed to me that the performance was an inherent part of the writing process; they got to write in public.

I told Julie about this discussion, and she asked, "What does Patrick think about writers selling manuscripts?" I presume you have no objection to preserving manuscripts, including editors' marks, galleys, and revised galleys?

Patrick Nielsen Hayden ::: (view all by) ::: January 20, 2004, 12:14 PM:

Why would I have any objection, either to writers selling their marked-up manuscripts or to you making those changes to your blog? This is about what I'm comfortable with, not what anyone else should do.

Personally, when I play music without an audience, I say I'm playing; "rehearsing" and "practicing" are words I apply to rather more goal-directed musical efforts, such as trying to learn a particular piece or running through a repetoire along with an ensemble. But never mind. My own musical abilities run strongly in the direction of improvisation and I have very little stage-fright or anxiety about making minor errors while performing. The point is that I don't like the feeling of being "looked over the shoulder" while writing. Is that really so odd?

Mitch Wagner ::: (view all by) ::: January 20, 2004, 01:53 PM:

Patrick - I wouldn't say it's either odd or not-odd; I'd say it's just different from the way I think, and so I found it surprising.

Perhaps the difference comes from our different professional backgrounds - my first real editor was a guy who was in the habit of saying, "Send me whatever you've got written on that story, I want to see what you have so far."

I'm used to the idea of having people look over my shoulder while I'm writing. Sometimes they'd literally be looking over my shoulder (as in, standing behind me and looking at the screen while I wrote). I only worked in that environment for four years, and it was 15 years ago, but it was an extremely formative environment for me.

P.S. And now here I am reading over this message in the preview screen, and I see that I'm getting an error message at the very bottom of the screen, underneath the previous messages in the thread:

MT::App::Comments=HASH(0x810adb4) Use of uninitialized value in sprintf at lib/MT/Template/Context.pm line 1187.

Patrick Nielsen Hayden ::: (view all by) ::: January 20, 2004, 02:22 PM:

Well, there are all kinds of things that I'm "used to" without particularly wanting them to become part of what I do for pleasure. I can write book jacket copy in front of a crowd, but it doesn't give me the same pleasure-of-composition that I get from, say, getting an Electrolite post right at my leisure.

That error message seems to be happening for a lot of people who've upgraded to MT 2.661. It appears to be harmless. There's a thread about it over on the MT support boards and I expect it'll be fixed by and by.

Dave Slusher ::: (view all by) ::: January 21, 2004, 01:39 PM:

OK Patrick, you seem convinced that this RSS diffing in NetNewsWire is a big deal while when I tell you that the same thing can be done with HTML you think that is an abstract and distant possibility that need not be considered. I took a few minutes (around 30 or 40) and wrote a program that will do a similar diffing of an arbitrary HTML page.

Go to this page:


and when it asks you for an URL, give it the url of my post you wrote back to yesterday:


I've made a few minor edits to it, which you'll see highlighted. I'm not planning on leaving this tool up indefinitely (there are security implications, frex), just long enough to prove to you that this thing you are worried about is not unique to the RSS and is not in itself worth crippling your previously highly useful full text feed. Any number of people could duplicate this in the same 30 minute time period if they wanted to.