Wheeling, Illinois.

Sportmart was a pioneer in the big box retailing of sporting goods and sporting apparel. Their stores were located across the Midwest, California and Canada and the corporate office was located in Wheeling, Illinois.

In my final months at Handy Andy, Sportmart was looking for a new CIO so I applied for the job. I didn’t even get an interview for the position even though I was a perfect match for the job. We had the same hardware, the same software and they were a similar size business as Handy Andy. I had been out of work for November and most of December when they gave me a call. The Director of applications development had resigned and the call was out to the recruiters to fill the position. I was a great fit. I was an expert on the hardware and I knew the retail software like the hair on my chin. To this day I thought they forgot that I was coming in for the interview. The Director of Applications Development had reconsidered and she took her resignation off the table.

I interviewed for the non-existent position and there was no job. The new CIO and I hit it off real well so he created a job for me. I became the Director of Technology Services in charge of all data center operations, networks, store point of sale systems, store support, the helpdesk and all communications. I had everything but AS/400 development but I had my work cut out for me. It was a great job with a great company and I loved the people.

Here are a few highlights of the challenges and comments. Some of the line items are technical (sorry).

AS/400

• 1. Reclaim storage had not been run in over two years on the main AS/400. This impacts disk storage, performance and management of damaged objects on the system.
• 2. QSPL, QBATCH, QCMN and some other production sub-systems were running their jobs in the *BASE pool. This greatly impacted the performance of the machine (very poor).
• 3. The operating system PTF’s (patches and updates) were years behind.
• 4. There was difficulty in upgrading the operating system and performing the reclaim storage because a temporary patch had been placed on the operating system a year ago, and it had been forgotten after the permanent fix was applied.
• 5. There were two AS/400’s. The operating systems were at different PTF levels. The development AS/400 operating system did not match the production AS/400. This makes testing invalid.
• 6. The system was not backed up. Recovery of the complete system was impossible. Only major physical files were backed up (none of the logicals or access paths). I put together a backup strategy to address this problem.
• 7. The disaster recovery plan could not recover our system. There was not enough disk storage, proper tape drives and communication to the stores was not provided for. I put together a plan and we fixed this problem.
• 8. Physical and logical files were in separate libraries. This makes recovery very difficult and time consuming. I reconfigured the files to repair this problem.
• 9. The system reply list had entries that dumped programs that had any problems. Programs would dump because of record locks. Instead of fixing one error, there could be hundreds of program dumps to repair. I couldn’t do anything about this because the of philosophical differences in the development area.
• 10. Batch processing was not managed. Long running, resource intensive jobs would run during the day which impacted the performance of more critical batch and interactive jobs. I created job classes which directed these jobs to night job queues.
• 11. The I.T. people moved jobs around and changed their priority at their whim. By reorganizing the jobs and queues, I was able to put together standards which would not allow anyone to change a job.
• 12. There was no third shift and no system set up to notify anyone if a problem occurred with processing during the night. I brought in a monitor system that would page the developers or the operations group when a problem occurred.
• 13. The credit authorization system that communicated customer credit card and check requests from the stores to the host had a design flaw. If the credit card or check company had a system or line problem, both services became unavailable. Fixed.
• 14. The credit authorization programs would lock up, usually on a busy day, and make the credit and check system unavailable for 30 to 60 minutes. This was due to bisync communication to an old IBM Series1 processor at our bank. We changed the protocol to SDLC and directed the authorizations to a different branch. At least half of the stores had to remain on Bisync because of a chip problem in the feature ‘T’ board on the 4683 registers. This was fixed for half the stores, the other stores had to wait until the lease ran out. The treasurer re-signed the lease without talking to the CIO or me.
• 15. Of the 250 GB of storage, at least 20% was garbage that had not been cleaned up over many years. There were work files, copies of files, test libraries and products brought in for evaluation that had never been removed. I worked through these lists with the developers and I got it all cleaned up. We recovered 30% of the storage.
• 16. Output queues were not cleaned. There was hundreds of output files in the queues, many of them were over 15 months old. This also impacts performance. Additionally, the QTOTJOB system value did not take these into account. It was set for 2500 and the number of jobs in the system was around 9000. QADLTOTJ was set to 25. This caused system stalls every time the 26^th report kicked off. I changed the settings and I got most of the queues cleaned up, but there was a packrat mentality that I had to work through.
• 17. Upload and download capabilities did not exist. The yearly store budgets had to be keyed manually into the system. It took an additional month to identify the miss-keyed information. This is a standard feature. I installed it and we got the budgets done in fifteen minutes.
• 18. The UPS system could not handle the load of two AS/400’s, the PBX’s, the networks and the voice mail. If the building lost power, the operators had less than one half hour to bring the systems down in a controlled manner. This was not possible and it would cause many problems. I replaced the UPS.
• 19. New disk storage was continually purchased instead of cleaning up the system. I cleaned up the system which recovered storage.
• 20. Programmer tools were on the production library list for interactive and batch processing. You don’t do this. You don’t want your programmer libraries available to your production programs.
• 21. Hawkeye, a programmer tool, was on the production box. That was because they did not update their test system. Programmers secretly abandoned the test system and they tested on the production side against files in their personal library. I put together a plan to refresh the test system and apply the changes every evening. The programmers did not want to work in the test system because the test files were only a 20% sample of the production files. They did not know how to test and it was out of my control.
• 22. The startup program automatically brought up all of the Sportmart production systems. The IPL Process could not bypass this procedure. If the option to bypass the startup routine was selected, the system console could not be brought up. I fixed this problem.

• 23. The stores polled the home office instead of the home office polling the stores. This would cause stores to miss important download information because the host applications might be in the process of building the nightly download files. This created a series of problems. I put together a plan to reverse this process and we were in the process of testing it when Sportmart was sold.
• 24. There was no price override report for the store managers. How could they manage their cashiers? I worked with the vendor and created a series of cash management reports. The home office wanted to see these reports so I automatically sent a copy to the home office during polling.
• 25. Communication to the host for credit authorization was interrupted continually due to an old version of communication manager on OS/2 and perpetual problems with the IBM 2218 FRAD units. This was fixed.
• 26. Maintenance contracts were not managed. There were to many maintenance vendors and it was not clear who maintained what. Some equipment was on maintenance with more than one vendor and some equipment was not on maintenance at all. This was a real pain in the butt. I’m not sure I found all the problems. I documented the maintenance vendors and labeled each machine and added it to the standards manual.
• 27. The point of sale systems had no disaster recovery. We replaced the tape units in each store.
• 28. The point of sale systems had unreliable prices. The POS system required a micro-managed price system. Because the prices were not micro-managed, permanent and temporary price changes wiped each other out. I designed a way to fix this problem and I repaired several other problems at the same time.
• 29. There was no historical return price in the system. If a customer bought an item for $1,000 and returned it without a receipt after the sale, the customer would get the regular price ($1,200). This was repaired at the same time I fixed the store pricing problems.
• 30. If a new item file had to be created for the store system it would not contain any ads or price changes. These had to be re-keyed. I designed a fix for this, the developers wrote it and we installed it. An untested bug sold two Big Bertha $250 clubs for $25.
• 31. There was no lab to test POS changes or bugs. I bought extra equipment for a lab.
• 32. Because there were two hardware platforms at the stores (4683 and 4694), any upgrades had to be coded twice, with twice the cost. We could not address this problem until the hardware lease expired.

• 33. The networks were unstable. There was OS/2, Windows NT, MacTalk and two versions of Novell. We converted everything to NT except the MacTalk which was it’s own island in advertising.
• 34. Sportmart invested in token ring and it was the network of the future. This added cost to the network hardware (vs. Ethernet). I directed us to go to the Ethernet standard for all future purchases. Token ring was going to be phased out.
• 35. There was very inadequate email. Email was handled through limited use through IBM Office vision. It was virtually non-existent. I was in the process of purchasing an email system when Sportmart was sold.
• 36. Backup of the Networks was performed through ADSM on the already over burdened production AS/400. I split the backups so the network backups were not dependent upon the production midrange.
• 37. Most of the 250 PC’s were slow 486 machines. Performance for power users was unacceptable. I put together a cost justification to replace a large group of PCs for the power users.
• 38. There were few diagnostic tools for the PC’s and networks. No line tester, voltage tester, port tester, scope, Lan manager, etc. My budget was approved an we purchased the tools required to do the job.

• 39. There was no management of the gift certificates and merchandise credits. It was unknown what was sold or redeemed. There were no serial numbers which would give an unscrupulous person the ability to make multiple copies. I identified the problem and we were in the process of fixing it.
• 40. Every week multiple copies of a 5-6000-page report was run and printed. This consumed a lot of paper and contributed to wear and tear on our laser writers. This report was necessary because the users had to go to too many different screens to get the information that was pulled together on the report. I submitted a program request to create a single screen for the users.
• 41. The host application systems are an old, pre-release version of JDA. We were going to upgrade to the new version. We didn’t have to since Gart Sports already was using it.
• 42. Several interactive programs had record locks on major files. Users were cautioned to get ‘in and out’ of these applications quickly or the batch applications would dump. This only needed a design change. Fixed.
• 43. Price change batches and Ad batches were automatically deleted after they were released. If a store missed the batch, the data entry group had to re-key the batch. Copies were made and place in a queue for thirty days.
• 44. Ad analysis was unavailable. Program request was made.
• 45. There was no standard price change form. Each buyer submitted a form of their own design.
• This was pretty funny. My data entry group had all kinds of problems because the buyers all had their own form. We standardized on one form.

• 46. There was no overhead projector. Slides were created on an old version of PowerPoint that had limited functionality. Once the slides were created, they were sent to a shop that created photographic slides. This required a two-day lead-time and cost an average of eight hundred dollars per show.
• This became a pain in the butt for me. I showed them how to use Powerpoint and I rented a projector for their quarterly meeting. Every time they had a meeting they wanted me to run the projector. They didn’t want anyone else to do it.
• 47. The electrical system in the computer room was very poorly designed. I had it redesigned and repaired.
• 48. Very little help desk support. The Help desk had not been established. I created a manger position and I promoted Mike into the job. (He did excellent).
• 49. A second AS/400 had been purchased with the intent to move some applications, such as E3, to gain performance for all systems. This project just died. I could not get this project aproved because it was a political football, even though we had an powerful second AS/400.
• 50. 100 Modems had been purchased for a dial backup system for the store’s credit authorization system a year before I had arrived. Like several other well-intentions and well funded projects this project just died. By the time we addressed the problem the modems had become obsolete.
• 51. The phone systems in the stores were in terrible shape. In some stores, as many as 12 phones were not in service. I was testing a new system when Sportmart was sold.
• 52. A RISC server had been purchased for Sportmart’s web site. This project just died and it remained dead. I had no idea what started this project but there was no owner so we used the system for something else.
• 53. A second leased line to our distribution center was available, but never used. We were getting billed for it. I brought it on-line to service half of the DC.
• 54. A second license for the JDA system was unnecessarily purchased for $64,000 for the test system. Sportmart did not tell JDA what it was for. There was nothing that could be done about it. JDA already booked it and the relationship between JDA and Sportmart had fallen apart before I arrived. I helped reestablish the relationship.
• 55. The voice mail system was far behind in upgrades. It was rarely backed up. Recovery would have been a long process. I replaced the voice mail system.
• 56. A portable phone system pilot was installed in one store and it was not managed. The supplier wanted his phones back. The store lost them. This was before I got there. A lot of projects got started and they became orphans. Not while I was there.
• 57. An auto attendant system was installed in a store and it was not managed for evaluation. The vendor’s equipment had been unplugged and there was no follow-up. I found the equipment and started the testing all over again.

[Home] [Professional] [Unicous Marketing] [Lumber Liquidators] [Gingiss] [Scarsdale Solutions] [Donald Bruce & Company] [Sportmart] [Handy Andy] [Claires Boutiques] [Recruiting Stories] [Weber-Stephen Products] [Family] [Drawings] [Stories] [Contact] [Personal] [Events]