Category: Technology

A colleague asked me this morning if I had seen the Microsoft Mesh announcement.  I said I had, and that I thought it looked really pretty, but I didn’t get it.  “It looks like a solution in need of a problem,” I said.  My colleague went on to say that this was basically what Joel on Software had also said, and I just got a kick out of reading Joel’s diatribe.  I wish I wrote half as well as Joel.

Although I am not excited about Mesh, I am not as pessimistic about it as Joel is.  I think it’s great that Microsoft has projects like this.  Microsoft should be trying to innovate in as many ways as it can; and with innovation comes experimentation.  That’s not a bad thing, although Joel seems to think it is.

My only concern is that this project appears to have come from Ray Ozzie himself. It seems like something that a couple of smart kids out of school could do.  If I had the resources of Ray Ozzie behind me, I hope I’d come up with something much bigger.  Products that don’t change the world are not worthy of the Chief Architect at Microsoft.  So far, Mesh seems smallish.  But, he’s a smart guy.  Maybe the world-changing part is still coming; and if so, that’s great.

Alright, this is just a prediction.

The decision for Microsoft this weekend is whether they still want Yahoo or not.  If they still want it, it will be worth it for them to expedite the deal and increase their offer.  Going into a proxy fight doesn’t help anyone, and just drags this out.  So, in my view, there is zero chance of a proxy fight; if Microsoft still wants Yahoo, they’ll raise the offer.

If, however, Microsoft has decided they don’t want Yahoo, they’ll walk away.  I don’t believe those that speculate Ballmer would go through with the deal to avoid “losing face”.  There are plenty that would think Microsoft did the right thing by walking anyway.

Now, should Microsoft walk away, the Yahoo stock will tumble; if it stops at ~$20, where it was before the Microsoft offer, Yahoo should be pretty happy.  It may drop steeper than that.  Yahoo management knows this.  If Microsoft really wanted to play poker here, they could just wait 6 months and pick it up for even less.  But Microsoft doesn’t want that.  Microsoft wants this deal done now so that they can get back to business.  Yahoo doesn’t want the stock to tumble.

Based on Microsoft’s public comments about the advertising market over the next 5-10 years, I believe Microsoft wants Yahoo.  They believe the market they are trying to conquer is far larger than the price of Yahoo.  They don’t want to pay more for Yahoo (who would), but they know that any price less than $50B still pays itself back over the next 10 years if successful (and of course, Microsoft internally believes it will succeed).  As such, they’re going to do everything they can to close this deal.

I suspect a nominally increased offer from Microsoft has already been made. This will be enough that the chief Yahoos can claim victory and accept the deal at the same time.  This thing is going forward.

But what do I know.  More predictions.

The Windows Disk Scheduler is simply retarded.  Here is what the usage of my disk looks like.  I’ve got 200GB of unused space, and Windows insists on fragging up the 16% of the disk that it actually uses!

If you don’t defrag nightly already, you should.  See this KB Article from Microsoft for instructions.

Recently eBay has been getting beaten up as it does fee restructuring.  It makes me wonder if an old idea is ready to come to life:  Distributed Classifieds.

Why Do We Need Distributed Classifieds?

When you’ve got something to sell, you should be in control. You have the goods, so you chose your marketing, you chose your price, and you chose your guarantees, service, and customer support.  Why should any company – eBay included – dictate to you how much they get?  Shouldn’t it be the other way around, and you tell eBay how much you’re willing to give to them?  After all – you are the seller!

eBay gets away with this because they are a centralized server.  They control everything, but also provide the product base from their sellers and the client base of buyers.  Since they created the market, you pay them to be part of their world.  Unfortunately, you sacrifice a lot of control and pricing when you do this.

Distributed classifieds allow multiple markets to co-exist and put the seller in control of the price.

How Distributed Classifieds could work – Finder’s Fees

With a distributed system, the seller would start out by posting his classified.  Imagine something like this:

Red Bicycle For Sale – $200
2 years old, good condition.
Berkeley, CA
Feb 20, 2008 2:13pm
Finders Fee:  $10

All of this looks pretty standard.  But what is the “Finder’s fee?”  The Finder’s Fee is the seller’s proposed “listing price” to any market.  A market is anyone that helps find a buyer for this product.  eBay, for example, could be a market.  Or, “John’s House of Bicycles” could be a market.  For whoever finds a buyer for this item, the seller is willing to pay $10.

Why is This Good?

In this new world, the seller doesn’t get locked into a single market.  With eBay, once you’ve listed your items you are obligated to sell through eBay even if a higher bidder emerges from another market.  For example, you couldn’t post your item both in the newspaper and also on eBay without losing reputation on eBay.  This is not right – you should be able to sell to the highest bidder under any circumstances.

Search Glues It All Together

One problem facing sellers today is which market to sell the products on.  Choices are numerous, but you want the ones which are most profitable, easiest to use, and have the largest collection of buyers.  Examples of markets you could choose include Craigslist, eBay, your newspaper, a piece of paper on your dorm bulletin board, etc.

But, shouldn’t you post it once, and have the markets find your listing automatically?  Just as search can find web pages, what if each market crawled the web looking for items for sale?  Instead of you going to the market, you simply post your listing, and they find you.  RSS has led the way here – using pingservers, you could post once, ping a few servers to let them know of your new posting, and then a plethora of potential markets would pick up your for-sale item.

Other Details

I’m leaving out lots of details about how reputation systems work, how to handle auctions, how to guarantee payments, and more.  Some of these problems are easy, some are hard.   But I believe they can be solved.

Once solved, the world is a better place:

  – Sellers rejoice in selling to the best customer. 

  – Buyers rejoice in getting lower prices. 

  – Businesses rejoice because there is no central system monopoly (eBay!) for all classifieds.

A distributed eBay would rock.

With the potential acquisition of Yahoo!, Microsoft could also pick up Zimbra, the deft little email company that has picked up a very large fan base in a short few years, but is now owned by Yahoo.  Since Zimbra customers are primarily those that rejected Microsoft Exchange, it’s no surprise that they aren’t very enthusiastic about this turn of events.

The best article I read about this was from Brad Feld, with the press release for the first Zimbra release after the acquisition:

Dear Microsoft Zimbra Customer:

The latest release of Zimbra (Zimbra 6.0) has been released to manufacturing today and will be available within 60 days.  As part of this release, we are renaming Zimbra 6.0 to Microsoft Exchange Server 2010.  The upgrade will be automatically delivered as part of your Zimbra license (sublicensed under the Microsoft Live Enterprise Support Service.)…

This is great news!

Since I have no insight into ads or Google’s financial numbers, I can post this.  I post this on my own, and have no idea what my employer thinks…

Last month, Markus Frind reported on his blog that his adsense click-through-rate “declined by 60% in the last 2 months”.  Content providers that make lots of money through pretty liberal placement of fairly useless advertising have been complaining.  It’s no surprise that they are disappointed; who wouldn’t be.

This week, Google’s 10K report says:

“…the main focus of our advertising programs is to provide relevant and useful advertising to our users, reflecting our commitment to constantly improve their overall web experience. As a result, we may continue to take steps to improve the relevance of the ads displayed on our web sites and our Google Network members’ web sites. These steps include removing ads that generate low click-through rates or that send users to irrelevant or otherwise low quality sites and terminating Google Network members whose web sites do not meet our quality requirements. In addition, we may continue to take steps to reduce the number of accidental clicks. These steps could negatively affect our near-term advertising revenues…”

Advertisers seem to be heralding the news (see comments), as you would expect.

This announcement is just really cool.

One site that seems to be consistently worth reading is Silicon Alley Insider.  Henry Blodget just seems to be the most rational and consistent writer around.  If you don’t read his stuff, you should.

If a company wanted to buy your company, and you’re not really excited about it, what would you do?  Of course, you’d just ask for a higher price.  If you end up bought, you feel good that you made the deal better.  And if you don’t get bought, you are still happy!  We should have known this is exactly what Yahoo would do!