The blog has been converted to the Google version. Hope it's better.....
Friday, December 22, 2006
Wednesday, October 18, 2006
I have been an eBay user since 1997. When I first saw eBay I thought it was one of the most ingenious ideas that had yet appeared on the web. It's self-policing nature, and its limited role bringing together sellers and buyers of all sorts of things, particularly antiques and collectibles (in other words, junk), I thought was brilliant. Since then a number of improvements have been made. With the integration with PayPal, the My-eBay one-page tracking center, automatic searches, feedback, and the ability to print postage, buying, selling, shipping and tracking it all have been made about as simple and easy as their are going to get. All the while eBay's popularity grew until a certain critical mass was reached. It's hard to imagine an other site getting into the auction business in any meaningful way. It's eBay that has the traffic, today. But one of the things I hope eBay will realize is that the critical component of this traffic is the buyer, not so much the seller. If eBay does not make some badly needed course corrections, buyers will go elsewhere. And once enough of them do, the critical mass is lost and will never be recovered. Somewhere, right now, some smart people are sitting around thinking up what a better auction site might look like. Some of them will, eventually, actually build it.
I'm guessing, but eBay probably sees the seller side as the principle revenue stream. I hope eBay will consider that it is the sellers that can most easily shift that stream to another auction site. They will follow the buyers. Buyers on the other hand, are needed in a large mass for the concept to work at all. Lose that mass, and sellers will vanish. eBay discounts the buyer's online experience at its peril.
- The search engine is buggy. Make it return correct and complete results.
- Add an option for "any of these words" or "all of these words".
- Simplify advanced search use.
- eBay provides any number of ways for buyers to find sellers, add them as "favorites" and view other items offered. Add the ability for buyers to exclude selected sellers from searches.
- Add the search selection on postage charged.
Make it work!
- Getting the web pages to function properly every single time must be the top priority. If it doesn't work, everything else is pointless.
- Cut down os graphics significantly.
- Remove references to slow and unreliable external servers, such as doubleclick.com, that reduce eBay's functionality and clutter the page, while doing nothing at all to improve the users' experience.
- Get out of unrelated businesses.
- Stop trying to compete with discount retailers.
- Remember to think of the buyers' experience, not just the sellers'.
- Focus on what made eBay a great service: online, person-to-person, auctions.
A lot of people are hoping eBay will change its course, and soon.
Tuesday, October 17, 2006
Something I thought was very clever about eBay when it appeared was the self-policing nature of the site. Such schemes are common around the web today, but eBay's was one of the first I'd seen with such features. The search and, particularly, the rating system go a long way toward providing a buyer with the confidence to use eBay. This was especially important in the early days. Lately, some annoying trends have appeared that the system does not provide a good, clear way to discourage. One of these is price gouging on postage. For example, I recently received the following from a friend and fellow long-time eBay user:
"I was looking for a PCI USBII card for a pc at work. They are pretty cheap - like a few bucks. But some places were charging upwards of $20 to ship
them. There were like 1341 listed when I looked so I had to wade through a few pages of listings before I found one for a reasonable *shipping* price."
This problem is exacerbated by sellers that do not fill in the shipping costs fields in eBay's listing framework. Dealing with this problem would be simple and quick without eBay having to use any heavy-handed requirements, by once again returning to the buyer community. It is possible to include the listing postage in your search results. This helps, but it is not the default. It should be. In addition to the ability to exclude specific sellers, eBay should make a maximum shipping cost an advanced search option.
Monday, October 16, 2006
It's the Web Stupid
Of all the things I'd really like to let eBay know, there's one that really stands out as critical to the site continuing on top of the auction heap: website performance. They can create all new "features" to nickel and dime us for, all new "services" and all the fancy graphics they want, and none of it makes any difference if the website simple doesn't work. I wish eBay would severely restrict all development of new ways to spam buyers with "bold" and "featured items" and "two line titles" until they where sure, and I mean dead sure, that the website itself performed perfectly all the time, and every time.
Not too long ago, eBay did reduce and simplify the number of images on their pages. This was a positive step. However, the fact remains that all the flashy graphics and logos add absolutely nothing to the buyers' experience, and add nothing to the site's ease of use, while serving to create longer page load times and opportunities for errors.
IIS server errors are also quite common. I can be browsing along, go to make a bid, and find out all I'm going to get in IIS page not found errors. Various combinations of logging out and back in, and clearing cache and cookies restores the functionality - sometimes. Other times, it doesn't and I'm simply done for the day. At other times certain items will become unviewable, or even parts of a page. I've also seen pages become malformed in various ways when sections of HTML are not transmitted for whatever reason. IIS? Please eBay, it's 2006. It's long past time to switch to some enterprise-grade server software.
from two weeks ago has not been paid for. The buyer did go through the eBay check out right away, but no payment has been received. The buyer had a zero rating, which I've seen before. But now the buyer's rating is -1. The negative feedback reads "DONT SELL FAKE ID.....NEVER PAID GOT FREE WATCH FOR NOW WHITE COLLAR REPORTED" on eBay item #120035097771;
I wasn't planning to post a watch this week. But if a couple more days go by, I'll repost #170033195420.
Thursday, October 12, 2006
There are some excellent search engines out there on the internet. eBay's is not one of them. Yes it's true that complicated searches using words included and words not included can be done. It is also true that criteria regarding time remaining, item location and payment methods can be included. But creating a search that will really narrow down to what I am looking for is difficult. Documentation is not readily available. And the specific ins and outs of it are frankly more than many potential eBay buyers are interested in uncovering and learning. In addition, there are things that, as far as I know, can't be done, like "any of these words" verses "all of these words".
The search engine will be most users first eBay experience. Shouldn't it be really, really good? Shouldn't it be simple to use and take users right where they want, without the clutter of "featured items" that have simply paid eBay to be there?
There are some excellent search engines around that have proved what a good system can do. And that is not the business eBay is in. eBay really should consider partnering with a major search company to index their site. Features such as the ability to filter out certain sellers and categories, filtering identical items, "all these" or "any of these" conditions are badly needed.
There's a repeating theme developing here. Put the buyer first.
Tuesday, October 10, 2006
Get With The Program
One of the trends on eBay most annoying to regular eBay buyers is the ever increasing tendency of sellers to evade, ignore and otherwise not participate in eBay itself. eBay has done a great deal of work to make buying and selling easy from posting an auction, watching it's progress, paying for items and even paying for postage.
It's been annoying enough to encounter sellers not using PayPal, but there are now sellers who refuse to do anything at all through eBay. There are sellers who will not even respond to eBay messages. Sellers need to realize that buyers will, given the opportunity, avoid buying through third parties that will not integrate with the eBay checkout. Every time, as a buyer, I have to enter my shipping address an extra time, or even worse, create yet another account on some other website, I make a note never to buy from that vendor again.
There was an improvement made recently where eBay began listing the shipping amount in search results. This allows buyers to avoid sellers that have not fully filled out their auction information with the postage. That was a good start. As a further service to buyers, eBay needs to create a way for buyer to block selected sellers from their search results and category listings. eBay, please put the buyer first.
Monday, October 09, 2006
It's Not Macy's
One thing I have always loved about eBay is that it's like the world's largest garage sale. It's wonderful to see so many individuals buying and selling, well, just anything. eBay proves beyond any doubt that one seller's junk is another buyer's treasure.
Recently, eBay's front page is plastered with colorful ads that look straight out of fashion magazines. These are obviously the wares of on-line mass-retailers. However there are already a very large number of sites on the internet available to purchase goods from ordinary retailers, and a large number of sites available for comparison shopping (not to mention physical storefronts).
The vast majority of these sellers use reserve prices or buy-it-now prices set a few dollars above the opening bid. Clearly there's not going to be any super deals to be found here. In reality, they are simply selling goods at a fixed price. So what's the point? eBay has entered into an extremely crowded space, to which it brings nothing new, at the expense of what is unique to eBay's business.
I can imagine a response to this; eBay is an open market and the money coming into eBay from the sales of full-blown retailers is as good as anyone else's. I hope eBay will be cautious about this. An email inbox is open to whomever sends email to that address, but that doesn't mean we don't object to spam.
Major retailers spoil eBay. They do not offer real bargains, they do not fit their payment and checkout systems into eBay's framework, and they are frequently unresponsive to email.
This week's eBay sale is 18 size Elgin pocketwatch. This one runs quite well. It's nice due to its age (1887) and really strong running condition.
Tuesday, October 03, 2006
Wednesday, September 27, 2006
This week's eBay sale is 16 size Elgin pocketwatch. 16 is the next size smaller from the 18 size watches I've sold recently. This one runs pretty well. It's a 1920s model with a metal dial that was popular at the time. These watch dials have not held up over the years as well as the earlier porcelain ones have. They are also difficult to clean. This on is a fair example, I've sure seen worse.
Monday, September 18, 2006
Thursday, September 14, 2006
Tuesday, September 12, 2006
Wednesday, September 06, 2006
How to tell when your technology project is already a failure
Nearly all business sectors are littered with the spectacular failure of
large technology projects and tech-related business ventures. In
hindsight these projects frequently seem to have been doomed all along,
and yet the course toward the inevitable disaster remained locked on.
Here we examine a short list of tell-tale signs that a technology venture,
organization, or large project is serious trouble; doomed to failure.
The most obvious symptom of an organization in serious trouble is staff
turnover. But not always in the most obvious way employees leaving the
company after shorter and shorter durations. A more subtle sign is what
occurs when a large project persists for too long with an unrealistic
schedule or requirement list.
The most experienced staff will almost immediately recognize the most
serious issues with a requirements list, the most serious bugs, and the
areas of greatest unanticipated risk. But the time at which fundamental
problems would normally be recognized and addressed will have long passed
as the project enters its most desperate and frantic stages. Staff which
understands issues and voices concern are normally part of the solutions.
In fact that is exactly what defines these individuals as experienced
veterans; they have solved difficult problems and know what to do.
normal project, discovering and resolving challenges is exactly the goal.
But when things are really going deeply and seriously wrong, something
changes. Normal development processes break down and, when an
organization is in trouble, experienced voices are not seen as
constructive. Senior individuals will be relegated to routine or
insignificant roles, replaced, let go or otherwise removed from the scene.
Newer and less experienced staff will move into their place. Such
individuals will have the can-do attitude management will be looking for
at this point, but this won't mean the work actually can be done.
Mistakes will be repeated and wheels will be reinvented in great numbers.
In a normal project these are considered risks. But in an organization
that has become truly dysfunctional, reinvention from the ground up will
actually be desired and encouraged. Meanwhile, the fully valid concerns
of the senior staff will invariably become part of phase II.
The organization has been engaged in frantic struggle to accomplish major
projects. Months and maybe year have passed yet unresolved design
questions, critical bugs and to be determines litter projects' status
reports. At this point management will begin to speak of Phase II. What
is happening is that failure has become so obviously unavoidable that a
way to establish a point of victory is desired. A subset of functional
requirements that appear to be achievable will become the focus, and other
issues will be pushed off to next year, or some similar mythical time when
everything is settled.
There are several problems with this. First of all, this is really a form
of denial in disguise. On the surface it may seem reasonable to scale
down expectations, create a near-term point at which victory can be
declared and table certain hard problems for later. This would seem to
allow the the staff to work with better focus and more breathing room.
However this is really a sign that certain aspects of the work are indeed
very difficult or have already spiraled beyond the grasp of the staff and
management. These very aspects of the work will nearly always be exactly
those that are critical to the project resulting in any meaningful interim
Employees know this. They know exactly what aspects of a project are
critical. And they understand that the creation of phase II (or three or
four) spell ruin for the artificially declared interim victories.
The second problem with this phenomenon is a fallacy also related to poor
time management; the idea that the future will allow more time than the
present, since more of the tasks on the table at the present will be
completed in the future. In other words, this is the idea that there will
be more time available in the future and so some tasks can safely put off
to a mythical future when there will be time. Of course what really
happens is that new tasks are adding as time passes, leaving the net
workload constant. Phase II will never come.
The worst of worlds occurs under the pressure of the death march project
when the very tasks that are both critical and difficult are put off to a
vague future time that will never actually come to be. Time will be
wasted in the present building work-arounds for problems that have been
avoided, and the interim work is nearly assured to be unusable.
Bad Becomes Normal
Bad becoming normal is exactly what is happening as an impractical
workload persists and the future, where time allows phase II to begin,
never quite arrives. An unsustainable situation, that management somehow
forgets was supposed to be a temporary condition, becomes simply the way
things are day to day. When this happens, management will come to believe
that the project's state can maintain its level of effort as ordinary
work. This would be nice, but the catch is that the longer this situation
persists the greater all the problems explored here become, and a problem
of scale develops.
If bad becomes normal, then the judgment of project managers is colored by
the desperately abnormal state which they have actually grown accustomed
to. Really critical problems will start to seem like ordinary problems.
Ordinary problems are perfectly safe to put off to phase II. Or to, for
now swept under the rug.
The Bulge Under the Rug
When things are really going south the last thing management wants to hear
is that a significant flaw exists in the current course. A really large
project, as it plays out, is bound to turn up significant flaws in
third-party software, unexpected information constraints at key points in
a process, unexpected license costs and other problems, large and small.
And these problems will be ignored.
The strange thing about this is that the software business is accustomed
to such issues, normally. In any normal project, time is alloted to
resolve such problems in advance. Even informal project planning will
including a certain amount of padding in the expectations. However we are
not talking about typical projects. In very large, mission critical, long
term and especially complex projects, the importance of such speed bumps
tend to, erroneously, be diminished in importance in view of the vast
landscape of the work. They are swept under the rug.
The problem with this is that just because a project is large, it is no
less vulnerable to the smallest problem in any component. And the more
bumps are swept under the rug, the more stressed the project will
ultimately be, and the more certain failure becomes.
Under New management
Nothing obfuscates problems like hiring new outside staff, particularly
management, to supposedly fix things up. The effect of this is an
automatic fog of confusion created as a result of new managers members,
with no historic knowledge; Rehashing already tried and rejected
approaches, Asking the question about already documented issues,
Introducing new, and almost always inferior, tools, misstating and
confusing the project's problems and status up the chain of command.
They net result of all this is nothing more buying time. The organization
collectively postpones having to deal with the reality of the situation
while the new staff comes up to speed. Problems that fall out are easily
attributed to a learning curve rather than to fundamental troubles
actually ailing the organization. At least for awhile... Eventually, in
many cases, the new staff is scapegoated for the projects failure. And
meanwhile, precious time is wasted that could have been spent on finding
real solutions to real problems.
All of these scenarios have common threads, and those general symptoms of
impending failure are another way to look at them. They all involve a
sort of procrastination, buying time, and putting off the inevitable
through delaying tactics, smoke and mirrors, red herrings and other forms
of fog. In addition, perhaps the most interesting point to make about
these situations is that, to one degree or another, those in decision
making positions must realize the gravity of their situation in order to
be motivated to set up these scenarios in the first place. This is
remarkable. Projects have found themselves in the state of inevitable
failure for a very long time. And we know it. And at some level we know
it while failure is occurring. And yet, we do nothing.
Wednesday, August 16, 2006
Monday, August 14, 2006
Sunday, August 13, 2006
I'm trying to organize, save, identify and HTML-ize photo collections
in a way that can stand the test of time. The reqirements for software I
can use are:
The information entered into the software for each picture or set of
pictures has to be stored in an open and accessible form, XML preferred.
No databases. No "importng". All the software required has to be
available 20 years from now (hence data sould be in an open and simple
The software has to create stand alone, simple, HTML files with all the
information in there and nothing extra required.
Running on Windows is a plus so my aunt can help typing in things.
I can't have the originals touched and the file names have to be
maintained as there are.
A nice way to browse and organize photos is also a plus.
The best tools in line so far are Google's Picasa, kalbum, picfolio, and
gwenview. I rejected others quickly for one reason or another. There are
others around also...
This is really a photo browsing program that's slightly better than what's
usually built into a file manager. In fact I think KDE using, or can use,
gwenview as a plug in? gwenview can generate standalone HTML albums with
comments and descriptions entered.
I can't tell where to enter comments and descriptions. It seems to be
looking for special files someplace, but it's not documented well or at
all. I also can not see how to customized anything it creates nor search
images by their descriptions and tags.
I thought this one would be perfect. It is a command line program written
in C/C++ that scans images and creates XML based on the meta data in the
files. It extracts the data put their by the digital camera including
exposure and flash setting, for example. It then uses XSLT and an
obviously alterable stylesheet to make HTML files. Great!
One would think that one could edit the XML and the XSL and get different
information into the HTML. You can't - picfolio always recreates the XML
from the images each time it runs. I could still use it to create base
XML if I wrote my own program to generate the XML, I would like it to be
in Java anyway. But that gets pretty annoying. Also, picfolio appears to
no longer exist. Its web pages are gone and I don't know that I can find
the source. And it's isn't a photo browser.
This program generates plain HTML from files you tell it to use. Source
files can be anywhere and it doesn't mess with the file names. This is
nice because the files can be stored in a directory structure that is
different from the HTML albums. You can create albums that re-use images
(picasa does this too). Notes and descriptions can be entered and they
are stored in a plain text project file. It's not XML but it's editable
and could be transformed.
The software is clunky and a bit buggy. It needs more GUI work. But the
main problem is that it stores notes and descriptions with the album
project and not with the photo file. Every time you go to use a picture
in a new album the information previously entered would not be there.
Yes, I could create more utilities that save this stuff and copy it into
new album projects... sigh...
I like the interface to this program. And it is the only one to runs on
Windows (oddly, it uses a built in wine interface layer to run on linux).
It stores notes and descriptions with the images so they are always
available even when you create new and different albums using the same
images over. It is an excellent standalone image browser and searches the
notes and descriptions. The HTML it creates it good too.
I can't figure out where it is storing the notes and descriptions. To
work for me I must be able to get at that information outside of the
application. Also it doesn't happen to run on my desktop, just my laptop.
Worst of all, it has to be able to see all the files it knows about when
it starts up. You can't record information about files that are on an NFS
mounted volume and are not always there for example. It does have some
way to "export" images (like to a CD) that may save the info
someplace, but I have not tried that.
So what to do? I could use picasa, by far the best photo browser, and
maybe put all the photos on a USB drive so they are always there. The
HTML albums it creates could be archived. But where's it storing the
information? And I could use kalbum to make HTML since it stores the info
well and doesn't need to access the files full time. But I can't share
typed-in information between two programs easily and kalbum doesn't even
store info with the images themselves, just with the HTML projects.
I could use picfolio the create XML files that include camera settings,
that could be helpful anyway...
No perfect solution so far has been found... I'll probaly have to write
some supporting scripts to merge and transform information. And getting a
solution common to Windows probably isn't going to happen.
You'd think there'd be a way to do this.
Wednesday, August 09, 2006
This is interesting. Borland has finally come full circle. They are bring back "turbo" line of products.
I was able to become a programmer because Turbo C was affordable to buy. And it was an outstanding product; stable, fast and easy to use. I even used Turbo C++ to write code for UNIX systems. It was a HUGE productivity boast.
Dropping pure C/C++ and their simple marketing style and product stategy was a huge mistake for Borland. Oh and changing the company's name too. What was that other name? I can't even remember...