floating atoll

Devaluing the practice of sniping auctions

eBay has developed a serious problem with sniping; the current response is to develop a better sniper. I propose a revision to the end-of-auction process, such that sniping becomes pointless.

Sniping is the art of registering a slightly higher bid five seconds before the end. It's a new thing in auctions; in most real-life auctions, a snipe bid results in the opportunity for a counter-bid; on eBay, a snipe bid can take advantage of the timeline and win.

Introducing a counter-bid opportunity into nearly-closed eBay auctions, then, might counter the problems introduced by sniping.

When a snipe bid is placed, a few seconds before the end of the auction, extend the auction slightly; provide a short period of time for counter-bids to be registered. Limit the bidding to those who previously bid on the item -- after all, the auction's closed at this point.

Each time a counter-bid is placed, extend the auction by a short period of time. This gives the bidders who've participated in the auction a chance to recover from a snipe bid, while keeping the auction closed to new bidders.

One possible scenario that could result from this is the death of snipe bidding: what's the point in sniping, when they can just outbid you? The act of sniping itself creates an opportunity for the snipe to fail; when you provide the other bidders a chance to beat your price, the value in sniping begins to fade, somewhat.

You could add restrictions on counter-bids, if you liked. If you post a counter-bid that fails, no more bidding for you until someone else beats the current bid; then you can try again. This would prevents malicious price inflation, while limiting the counter-bid process to active, interested bidders.

The end effect is to circumvent sniping entirely, to make it an ineffective way to bid. That'd be nice to see; I haven't purchased anything from eBay in over a year because of sniping. Perhaps they'd get some other disillusioned customers back, too.

Update: One possible solution would be to mask the highest bid towards the end of the auction, such that people could see what had been bid recently, but couldn't see the maximum to snipe it — or to make intelligent pricing calls. It's an interesting theory.

Drawing your attention to feedback.

I found Caterina's site recently, and discovered a really interesting article. It talks about "bringing things to people's attention", which resonated deeply with something I do a lot of: providing feedback.

I try to provide feedback on everything I see that can accept it; burnt out traffic signal lightbulbs, typos on corporate websites, articles I read. I guess it could be taken as a way of bringing my opinion to people's attention :)

Once I called Budweiser because my friend's box of beer had half the labels put on upside down. After the QA engineers stopped laughing, they sent me a $25 cotton tee shirt -- not a cheapo Hanes, but a REALLY EXPENSIVE well-printed cotton t-shirt. You can use this thing as a towel, it's that cool.

A few days ago the traffic signal at a local, heavy-traffic intersection was out. I called 911 and reported it, since it was about bar-closing time -- and the locals have a bit of a drunk driving problem. Didn't want to see anyone get hurt, and 911 knew what to do with my report.

I take a small percentage of the articles I read through in my news feed list (currently at 113 feeds, yay) and forward them off to various people. Two of my friends are specifically interested in linked-topic/semantic-indexing/slicing type technology, for their new mail application; another is interested in anything semantic-web/idea-map/brane-grok type stuff, and he gets a different collection; I share wireless-related links with Schuyler, the author of a popular wireless gateway program called NoCat.

I use my blog to bring my opinions and ideas to people's attention, which seems to be a worthy use for a blog; ideas like this one, for instance (meta-recursion!) are brought to my attention by the newsreader, and I have a chance to survey them, allocate them into mental buckets, and perhaps bring them to other people's attention.

Perhaps that's the killer app of blogging; it lets you bring something to the attention of people far sooner, far quicker, without economical or distribution barriers. Anyone can bring something to the attention of all who will listen, if only they care to speak. Email had this feature as well, but it's hard to link to a specific email on the web. With blogs, you can use permalinks and TrackBack and their ilk.

Tonight's show is about community-driven radio stations.

Recently, Doc Searls suggested that a use for internet radio could be to broadcast news-like talk shows, generated from posts on blogs. Kind of like All Things Considered, with listeners from your blogroll. Then Mitch pointed out that he's got some of the resources required to make this happen, and is looking for energy. I've some energy to spare, so I've transcribed an evening of thoughts here.

I've recently been searching passionately for a way to coordinate and assemble a gigantic, influential community of people; I think that this can help. There's a lot going unsaid to the government these days, at least not that they're influenced by; I'm searching for ways to change that.

I'd like to see a spider that maps the paths between all the different blog posts it can find, using Trackback and hyperlinks. Reads an entry, links over to somewhere else, reads that. Has a weighted preference for Trackback pings, and has a chance of staying at that page that goes up the longer it stays at a page, checking links, so it'll eventually just kind of trawl the whole network for news, in a connected fashion.

Give it the list of blogs from your favorite collection of blogs, and keep track of things that it finds that are blogs and add them to the list, too. Just make sure the blog's method of providing content is supported by the radio software. RSS feeds of excerpts, RSS feeds of full articles, a source, with three or four headlines, mix them together, or do waves of long articles, followed by short articles.

For programs like All Things Considered, you could pick a blog post and have it read all the trackback results from it in order. Or pick a liveTopics category, filter out the blog posts, and set it on random until it completes them -- or date-ordered.

I can record an hour show each day on a CDR, which makes a month of archives about $5. That's cheaper than cassettes, and I can listen to the morning radio show during the day -- and at the end of the day, I can go back to a specific entry and click on links I heard about.

Distribute a very cheap circuit board design for a micro-FM transmitter, instructions for assembling it either cheaply or in a waterproof container. It doesn't have to look nice, it just has to work. A house at a time, and I'd listen to blog radio in my house continuously.

Move the revolution to software that runs on the user's desktops: an application that watches for new articles. Use NNW to find new, unread articles, reads them aloud, and reads them as they occur. Pull the music from the user's active iTunes playlist, playing them a song at a time, pausing when necessary; all controlled via some master AppleScript that muxes them, following its own playlist.

Blog radio is one of those ways to improve communication and build community, and perhaps fix the radio world forever. Why listen to the big commercial station when you have better music (and blog posts!) waiting at home? Just tune it with your FM radio, and off you go. Adding in the blog aspect means that you, too, can make your thoughts heard, if you just care to make a blog.

Share playlists with other users; there's software now to uniquely identify music without concerning about file format, and you can share starting point in the article world pretty easily -- say, on your blog -- and congratulations, you just got your writing aired on the radio.

Thanks for listening.

Contribution: My Amazon Wishlist RSS feed

Recently, someone showed me the perl module XML::RSS. I've created a script that can be run on a daily basis to update an RSS feed containing the items in my wishlist.

It's turned out rather nicely, so I thought I'd contribute it to the Lazyweb. Formatting suggestions are welcome, I'm open to comments. Creative Commons attribution license in place. ISBNs are recognized as seperate from ASINs, and treated differently.

The Perl source code is available. I run it once a day; configuration options are at the top. It could probably be improved, but for now, it works, and that's enough.

Lazyweb: Centralized "wish to contribute" list

Lazyweb: I'd like a place to register my desire to support a given cause monetarily, even if I don't have the money. I can't give $5 to Doc right now (for his stolen powerbook), but darn it! I'd like to at least say "I wish I could" somewhere my vote will be counted at.

This would also be useful for noting my intent to register shareware; I use shareware apps until I can pay for them, and I'd be totally thrilled to list that I use them in the meantime.

It bothers me that I have to actually have the money in order to express that I'd like to give the money. It's important to express support in some form, even if I happen to be temporarily unable to, monetarily.

A liveTopics-like list of software applications attached to a per-user public or private application and amount list would cover it. It'd probably be best if it was a web-app, or maybe a MovableType blog with Categories (links into specific liveTopics) and the intended amount in the Subject (and NNW displays categories, so we're set there for subscribing).