Major Competition

[Parked]

Some of you may already know about this. It looks like major U.S. newspapers are getting competitive with their online classifieds, coming out with beta versions with a lot of features and relatively cheap prices, even offering free ads for $500 and less. Also offering auctions and buy it now as part of a premium package. After checking my main local competitor's site (a major city newspaper), it looks like my site may need major changes in order to be competitive in my local market before I've even launched. Know what your competition is doing.

[umlaw]

Im getting ready to purchase the Classified Enterprise edition and thought Id look through the forum. Youre absolutely right, Ive read a few articles lately stating that newspaper distribution is down because people get their news for free on the web now and hence newspaper advertising is down as well so their investing heavily in the internet and their classified sections because the major part of their profits happens to be coming from their classifieds. A major source of Yahoo's income is its classifieds as well- I think its more than 30% of their income. The more competetive they all become, the lower their advertising prices wll be and the more features they add to the sites to out do eachother- and us!
It seems to me that Geodesic is the best core application offered in the USA on which to build a strong classified website, but yes, I think we'll have to add a lot of 3rd party applications to add features that the Big Players have to compete and stay in the game

[mwester]

having used Geoclassifieds for 2 years now, it seems that the pace of development has slowed to a virtual stop. I think 2 years ago this app was cutting edge but it seems a dedicated programmer can build a more robust one from scratch in a few months that would outdo geoclassifieds.

[admin]

having used Geoclassifieds for 2 years now, it seems that the pace of development has slowed to a virtual stop. I think 2 years ago this app was cutting edge but it seems a dedicated programmer can build a more robust one from scratch in a few months that would outdo geoclassifieds.

Hi Mwester,

Actually, your assessment couldn't be further from the truth. We are in the process of completely refactoring the code for speed, documentation, ease of use, and security. This will allow us to produce more modules and additional features with less probability of problems. In fact, we have just hired two more developers this past week to help speed the process.

Also, if you know of a programmer that could outdo this software with only a few months of programming, please send us their resume asap, as we would be interested in talking to them.

[mwester]

i'm very glad to hear about that ... when is the next version scheduled to be released?

[Parked]

My purpose for starting this thread was to give a heads-up to checkout what your competition is offering with their sites, not to question the capability of the Geodesic software. I'm confident I can be competitive using their software and Admin just gave some great additional info I was'nt aware of. The changes I spoke of were mainly in the look of my site as much as anything. Also, I have'nt taken advantage of all the functions that Geodesic makes available.
Maybe I've said enough. Some of you may be my biggest competitors.

[McGrats]

I have to throw my two cents in here!

I've been studying the media, and in particular polling techniques, for the past 20 years or so. In fact I authored the much read "Hosing of America," a research paper on how the media has been "hosing" (manipulating) Americans through lies, distortions, and yes, polls.

The bottom line with the media's jump to the 'net is simply this: they long ago passed that arrogance threshold in which the thinking public considers them reliable. This led to the lost readership and resulting loss in advertising revenues. The same people who abandoned the Lamestream Media at checkout counters, will also shy away from them on the 'net. The media will attempt to puff up their website numbers in much the same way they did to their print versions, but I seriously doubt if they will ever impact a well designed website featuring quality items.

Auction websites fill a select niche and if marketed correctly, will be a success. I personally own several store fronts and my experience has shown one should not expect to begin making big bucks until you get established. I'm expecting the same for my auction website at www.northshoreauction.com .

Personally, although I'm just getting familiar with Geo, I've been able to make it attractive and distinct by simply planning ahead and aiming for a very select niche market. You can do the same! This is a relatively easily modified program, that can pretty much do what you want it to do (within limitations)!

Take car, and lots of luck!
8)

[russellmcox]

It would be very helpful if we know what was coming and when!

[mal]

We are in the process of completely refactoring the code for speed, documentation, ease of use, and security. This will allow us to produce more modules and additional features with less probability of problems. In fact, we have just hired two more developers this past week to help speed the process.


Great news! Keep up the good work.

[jaydub]

Actually, your assessment couldn't be further from the truth. We are in the process of completely refactoring the code for speed, documentation, ease of use, and security. This will allow us to produce more modules and additional features with less probability of problems. In fact, we have just hired two more developers this past week to help speed the process.

How much (if any) time has gone into optimizing the database schema for GeoClassifieds? I can tell everyone here that if you run a site with a decent enough size of traffic, geoclassifieds will fall flat on its face query-wise. At least as of 2.0.4 version. Our site once it reached a certain level of traffic started to experience serious delays. I dumped out all the queries geoclassifieds runs over a typical 5 minute site use period and then ran each query through the MySQL query analyzer. There are a number of tables in geodesic that really needed table indexes created.

After creating about a dozen new indexes across various tables, performance was dramatically faster than the stock geoclassifieds.

I would hope that the new design addresses some pretty questionable design in the current version such as duplicating the same code 20 times for 20 'optional fields' where only a couple of lines in each code block differ rather than a function that can iterate over 1 to N 'optional fields' with just one block of code.

Can we expect the seperation of code and display this time around? The ability to customize geodesic is greatly reduced due to the sprinkling of output inside classes and functions that are mixing processing of logic and display of output.

What about extending the number of HTML page modules or PHP page modules that are available? Having only 6 PHP modules is quite a limit on customization and especially integration with other website functions (say forums or blogs).

And finally what possibilities of discounted product upgrades are there? When inquiring about obtaining the 2.0.5 enterprise version the answer was that there is a 1 year period of version upgrades. Subsequent to that one must pay the full price for the new version? That's quite an upgrade penalty and in light of the comments expressed here by others regarding competition in the space, would seem to dissuade folks to stick with geoclassifieds.

anyways, thanks for listening to my hopefully consructive criticisms.

[smokey216]

Jaydub, some good points about the recurring code throughout the software. That's one thing that has really frustrated me...and I doubt anyone who didn't have some programming background (or at least HTML knowledge) could truly customize the Geo software. I love the software, just hope that much of the redundancy is addressed in future versions.

Also a couple other suggestions for the Geo folks--anything that is added should be able to be edited. It's frustrating to want to go back and change something like the display order or a category drop down question title, and to have to delete and readd instead. Also some of the pricing plans don't allow text input to choose a cost; instead you have to scroll through a VERY large list of drop down choices. I think text input would work better here to allow folks to set their own prices. Also not all numbers are available to choose from.

Might I ask, Jaydub, how you went about indexing the database? I am not too familiar with MySQL, just SQL, and mostly with stored procs and queries, not really indexing and optimizing.

Any help would be greatly appreciated!

Thanks much!

[jonyo]

There is some good input here. We have recently started making optimizations to the database in the latest beta software. We have also created tools to help track down slow-downs in the code, and optimize those parts of the code.

As for adding indexes to speed up the database: A general rule of thumb, is that if you see a query end in(as an example):

...WHERE some_column = value
Then adding an index to "some_column" if there isn't already one, will speed up the time that the query takes to run. Note that if the column already is "UNIQUE" or "PRIMARY", no index is needed. If you see an index and one of those identifiers on the same column, remove the index as it is un-needed and can bloat the size of your database. There are many more tips on how to optimize tables and speed up query time, a google search of "database optimize tips" or similar should bring up some good resources. Just be sure to ALWAYS BACK UP before you make any changes to the database.

[danzo]

Any area of improvements to output XHTML-compliant codes in the works?

[jaydub]

To examine a query in mysql, you can use the EXPLAIN command followed by the query in question.

mysql> explain select * from geodesic_userdata where communication_type=1;
+----+-------------+-------------------+------+---------------+------+---------+------+-------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-------------------+------+---------------+------+---------+------+-------+-------------+
| 1 | SIMPLE | geodesic_userdata | ALL | NULL | NULL | NULL | NULL | 30421 | Using where |
+----+-------------+-------------------+------+---------------+------+---------+------+-------+-------------+
1 row in set (0.03 sec)

This is just an example not one that I indexed. In this case the query will always scan the whole table of 30K+ rows even if you were just looking for a small subset of users (say if just 5K had communication type of 1).

For more info you should search online for the mysql EXPLAIN command.

[mwester]

We are in the process of completely refactoring the code for speed, documentation, ease of use, and security. This will allow us to produce more modules and additional features with less probability of problems. In fact, we have just hired two more developers this past week to help speed the process.

what's the latest on the update?