January 14th 2013 Posted at Uncategorized
I have spent some time re-factoring my old earlymiser/shopsort.com stuff. My first attempt at price comparison relied on a number of networks which created a maintenance problem. Maintaining that many web services was very little bang for a lot of buck, and it really didn’t serve the customer’s interests. Additionally it was a complete hassle getting users to set up pricing triggers as feeds. For those of you that don’t remember, you could set up an RSS feed for a specific product and then watch the price. Additionally you could also set up a specific trigger point for price and then have the application notify you. While that sounds good in the abstract, the UI wasn’t easy enough to use. There was also the problem that users’ don’t have enough information to make a reasonable trigger. Their prices were often so low, the trigger never got tripped. So this time I am effectively taking opposite track and using the web service more like a discount discovery. You simply select the top level category (click on the “Daily Deal” next to it), then select the percentage discount on the next screen. You can select a discount up to 80% and see what pops up. I am still wondering whether or not to go through the hassle of having the system generating RSS feeds that users can subscribe to. It seems to me it would be more useful to do a facebook app where the user would see what was the daily discounted deal in selected categories than generating RSS feeds. In any cause, poke around on it, ShopSort.com or EarlyMiser.com, it’s built on a single API (Shopzilla).
Update:Someone pointed out it might be helpful if I gave an example URL so people can understand how the discount discovery works. Here’s a link
Appliances at 60% discount. The system will then pull items that are discounted 60% or more off retail. The real sweet spot for finding deals is between 40-60%. At a 80% discount or higher you get a lot of items with artificially high list prices and low sale prices. To see different offers, simply select a discount level at the top center of the screen.
October 23rd 2012 Posted at Uncategorized
This is a PDF I built for a friend of mine. It’s the complete documentation for Python 2.7.3. I combined the 26 PDFs that are normally contained in the documentation into a single PDF. I then built new bookmarks to make browsing easier. So here it is. Please feel free to download it and use it if you like. Python 2.7.3 Documentation
September 26th 2012 Posted at Uncategorized
Thesourcebookonline.com is an online resource that offers sources for the entertainment industry. The website was once a printed book called Debbies book, founded by Debbie Hemela in 1978. The book represents every major prop house and resource for the entertainment industry in Los Angeles. Debbie personally inspects the inventory practices and companies listed in the book. This means prop masters, set decorators, and and set designers can get vetted sources for their productions. It is the most complete sourcebook for the entertainment industry. When selecting the best LA prop houses it is important that the prop houses be vetted. Designers need to know that a prop house is reliable and has the item they’re looking for. All resources in thesourcebookonline.com have been vetted to confirm that all items that are listed in our database are actually available. Additionally the sourcebook online indexes prop houses date databases in real time. This means you get the latest information for prop house. Debbie also runs a daily prop house blog which she updates with news about sources. For the set designer or set decorator looking for an edge, Debbies Book is the resource you need.
February 5th 2012 Posted at Uncategorized
The Super Bowl KickOff time is 6:29pm Eastern Time.
January 20th 2012 Posted at Startups, Uncategorized
I went to the Coronado Ventures Forum on mobile apps which only had one speaker with a mobile application. (Sorry Ipad only apps don’t count in my mind). Mobile apps are just that mobile apps. Ipad usage is similar to netbook usage by consumers and typically that is at home. That was the CEO of Jack Rabbit Systems, Andrew Van Luchene. It wasn’t a bad event. I wish the moderator had kept a firmer hand on the tax crank who showed up.
Yet another reoccurring theme of the night was the difficulty of finding talent in New Mexico. By talent people mean engineering talent. It was problem at Deep Web and it’s going to be a for any startup in New Mexico. My advice of course is to design your development environment to be entirely virtual so you can go after talent where ever you can. This requires a more disciplined approach in the design and requirement phase of the project as you cannot rely on informal communication to sort out issues. This means Github for code control, campfire for virtual discussions and some form of project management tool that is virtual. That will allow you to scale your organization without being tied geographic regions.
August 24th 2011 Posted at Uncategorized
After debating a number of people online who constantly claim that either the datasets or the climate model software isn’t available for download, I have decided to put together a comprehensive list of where you can download datasets and software. I know that in many ways this is a futile effort. Getting the data and models up and running isn’t a trivial task. In fact like a lot of scientific software it requires a degree of knowledge. After I sold I my company I spent 18 months looking at the issues, models and data. I am at best an educated amateur.
In any case here are the datasets and climate models.
- Surface temperature anomalies (GISTEMP (see also Clear Climate Code), HadCRU (alternate site), NOAA NCDC, JMA, BEST)
- Satellite temperatures (MSU) (UAH, RSS, Zou et al)
- Sea surface temperatures (Reynolds et al, OI)
- Stratospheric temperature
- Sea ice (Cryosphere Today, NSIDC, JAXA, Bremen, Arctic-Roos, DMI)
- Radiosondes (RAOBCORE, HadAT, U. Wyoming, RATPAC, IUK, Sterin (CDIAC), Angell (CDIAC) )
- Cloud and radiation products (ISCCP, CERES-ERBE)
- Sea level (U. Colorado, NOAA)
- Aerosols (AEROCOM, GACP)
- Greenhouse Gases (AGGI at NOAA, CO2 Mauna Loa, World Data Center for Greenhouse Gases, AIRS CO2 data (2003+))
- AHVRR data as used in Steig et al (2009)
- Snow Cover (Rutgers)
- GLIMS glacier database
- Ocean Heat Content (NODC)
- Ocean CO2 (CDIAC)
- GCOS Essential Climate Variables Index
- NOAA Climate Indicators State of the Climate 2009
These are weather models which have the real world observations assimilated into the solution to provide a ‘best guess’ of the evolution of weather over time (although pre-satellite era estimates (before 1979) are less accurate).
These is output from the large scale global models used to assess climate change in the past, and make projections for the future. Some of this output is also available via the Data Visualisation tools linked below.
Downloadable codes for some of the GCMs.
This category include links to analysis tools, simpler models or models focussed on more specific issues.
These sites include some of the above data (as well as other sources) in an easier to handle form.
Much bigger indexes of data sources:
June 24th 2007 Posted at Uncategorized
Managing your credit and your cashflow in a small start-up is sometimes difficult. If you are running short of cash cash advance instead of allowing yourself to get a late fee.
May 17th 2007 Posted at Uncategorized
I occasionally get people making offers on a site I have done. I recently got one offer for Early Miser that was particularly funny and I thought I would share it.
April 14th 2007 Posted at Uncategorized
When Shopping.com recently launched a product blog, Whatsinstoreblog.com it became apparent to me that blogs have a key merchandising role to play in online retail. It’s also apparent that this currently is an unfilled marketing niche for most merchants.
April 10th 2007 Posted at Uncategorized
Note to Google – Time to integrate the two tools I use the most at Google (except Gmail of course). I am pretty tired of logging into two seperate tools that really should be integrated or at least offer the ability to be integrated.