Today, Tyler R. Tichelaar of Reader Views is pleased to be joined by Roland Hughes, who is here to talk about his "The Minimum You Need to Know" series, which includes "The Minimum You Need to Know to Be an Open VMS Application Developer," 1st Impression Publishing (2006), "The Minimum You Need to Know About Logic to Work in IT," Logikal Solutions (2007), and "The Minimum You Need to Know About Java on OpenVMS," Logikal Solutions (2006).
Roland Hughes is the president of Logikal Solutions, a business applications consulting firm specializing in VMS platforms. Hughes serves as a lead consultant with over two decades of experience using computers and operating systems originally created by Digital Equipment Corporation (now owned by Hewlett-Packard).
With a degree in Computer Information Systems, the author's experience is focused on OpenVMS systems across a variety of diverse industries including heavy equipment manufacturing, pharmaceuticals, stock exchanges, tax accounting, and hardware value-added resellers, to name a few. Working throughout these industries has strengthened the author's unique skill set and given him a broad perspective on the role and value of OpenVMS in industry.
Mr. Hughes's technical skill sets include the following tools that enable him to master and improve OpenVMS applications: DEC/VAX C, DEC/VAX C++, DEC BASIC, DCL, ACMS, MQ Series, DEC COBOL, RDB, POWERHOUSE, SQL, CMS/MMS, Oracle 8i, FORTRAN, FMS, and Java, among others. Being fluent in so many technical languages enables Hughes to share his knowledge more easily with other programmers. This book series is an effort to pass along some of his insights and skills to the next generation.
Tyler: Thank you for joining me today, Roland. Would you tell us first what makes your books stand out from other books about Java and VMS?
Roland: For OpenVMS, that's easy. There are no other application development books currently in print for it. There are quite a few systems management and integration books out there for it, but none focusing on application development or even language usage.
As to Java, I did not drink the kool-aid in Java Town, and you won't find my body stacked in one of the piles being discovered there. I work with Java when I have to. It is not, and should never be the language of choice for anyone serious about application development. My book on Java dives right into the hard stuff: Calling system services, using run-time libraries, reading and writing RMS indexed files, interacting with the user on a VT-320 terminal. You don't find any other Java books talking about such things because their authors don't grasp enough about the language to accomplish it.
Tyler: You said Java "should never be the language of choice for anyone serious about application development." Why is that, and why do you think other authors have difficulty grasping it?
Roland: One has to define first "serious application development." While the WEB may become a serious portion of income for many businesses, it should never be serious application development. All of the serious application development occurs on the back end. We now call this SOA. You put a tiny little WEB service up which makes a secure call to a back end process that actually does all of the work.
Java is unfit for back end server development for the same reason almost all 4GL tools were unfit. They are interpreted. OK, they are p-compiled and that is interpreted. You cannot get enough performance, robustness, and security from an interpreted tool set.
If you look at most SOA implementations now, they are putting little WEB services up which communicate via some proprietary messaging system to a pre-existing back end which was written in COBOL, BASIC, FORTRAN, or some other language the trade press has long forgotten about.
Your question is its own answer: "Why do you think other authors have difficulty grasping it?" They are authors, not professional software developers. They are paid by a marketing war chest that has funneled money to one of the large publishers. The large publisher gives them a $4k-$5k advance and tells them to drink the Kool-aid with this book. They also tell them they have to put out 5 additional books this year per their contract. Exactly how much skill, knowledge, and research goes into any technology book put out by a large publishing house? Zero. They are busy churning out oatmeal for the masses.
When I wrote "The Minimum You Need to Know to Be an OpenVMS Application Developer" I took an unpaid year off to write that book. Had I been working for a publisher, that book never would have been printed. Assuming I was allowed to write it, the book would have been split into 9 different books, each one a watered down shadow of what the book I put out myself currently is.
Tyler: What do you think should be the language for application development and why?
Roland: That answer really depends upon your platform and the tool set you are working with. If you decide you only want to work with RMS indexed files, then hands-down DEC BASIC is the tool of choice. You must be aware that you have limited the size of both your application and your company by choosing to use RMS Indexed files rather than a relational database. Once a single indexed file starts spanning multiple disk drives it becomes very slow to access.
You decide, for whatever reason, a primitive relational database will be your data storage method of choice. You choose MySQL because it is free. You are limited to C/C++ as your development language on most platforms when using that database.
If you decide to use the best of the best in database technology, RDB on an OpenVMS cluster with fully distributed databases, you can literally choose any language supported on the platform, even Java as the Java book in this series shows.
In today's world, you choose your tools first: screen management, database/storage, messaging. Then you pick one of the languages that work with the tools you have chosen on the OS you choose to run.
Tyler: For the layperson, would you tell us a bit about OpenVMS and its role in the computer industry?
Roland: OpenVMS was and still is the most advanced operating system ever created by mankind. In the 1980's VMS gave the business world clustering and set the standard so high no other operating system has even come close to the implementation. There are a lot of OS's and vendors of OS's who will claim they have "clustering" but it is untrue. They have to spin a new definition of clustering, in most cases down to "we can spell the word clustering therefore we must have it." No version of Unix or Linux actually clusters. This is something Oracle is finding out the hard way with their RAC10 product and some much publicized travel site outages.
Were OpenVMS re-introduced today as a brand new operating system it would set the entire IT industry on its ear. Most of the IT industry is waking up to the fact that no matter how many $800 PC's you stick on blades, it is not a stable enough platform to run your company on.
Tyler: Roland, I must admit, I am not overly computer-savvy, and I find it difficult to communicate with IT people because of the jargon and the technicalities of technology. Therefore, I am surprised and pleased to meet someone who writes books about computers. What made you decide to be an author about technology?
Roland: It's the field I work in, and it is highly misunderstood. The industry has been reduced to 4-color glossies and MBA's making knee-jerk decisions based upon which product seems to have the most 4-color glossies in the press this week. We have to change that. There is a very troubling mindset in upper management that IT workers are just like the box stackers on an assembly line. This has led to a mad rush to off-shore IT and to flood this country with H1-B workers. Besides decimating the economy, these decisions are decimating business. From the 1970's through the 1980's a company's business edge was its IT department. This defined how your business ran and let you outrun your competitors. Now there is a trend to use the exact same software as everyone else. You no longer have a business edge, so MBA's enter a price war to outdistance their competitors. All a non-IT person needs to do is read the announcements from the SEC investigating accounting practices, stock options, and the rash of other scandals to see where price war mentality puts you.
Doomsday type people have been preaching we will eventually fight a world war in the Middle East over oil. If present trends do not change, we will fight a world war to get our source code and technology back long before we go to war for oil. Someone needs to put what we will need to recover from that war in writing long before it happens. They also need to point out that it is coming.
Despite what the off-shoring contracts say, many corporations no longer own their software. The data centers it is hosted in are in another country. If the owners of that center cut the network links, how does that company continue to function?
Tyler: Wow, Roland. I never thought about technology in that global of a way. What do you think is the solution to this situation? Is the situation something that companies need to solve for themselves or is government intervention required?
Roland: Businesses will not solve it for themselves. They have run headlong off this cliff and are too busy looking for another profitable scam that will let them avoid prison (like back dated stock options did for years).
Government intervention will happen, but not for any of the reasons you might think. Some incredibly large and stupid company (think Oracle or Microsoft) will have 70-80% of its source hosted on off-shore services (both of these companies have close to that in off-shore work now if you can believe the numbers floating around). At some point an entity or party with a fanatical national policy will take control of the government in that country and nationalize all of that source code. (Cuba did this when Castro took over, and other countries have done the same, so I'm not really stretching anything here).
Imagine what happens when those multi-million dollar Oracle products are no being sold as Alah-DB or some other radical name for $50.00/copy. Massive amounts of campaign funds get deposited to the re-election campaigns of all federal officials and congress declares war on the country that did this to protect Oracle (or Microsoft). Tens of thousands of your sons and daughters come home in body bags because corporations were both too stupid and too greedy to realize this off-shoring thing was a bad idea.
Take a look at GM and the other large companies off-shoring all of the software required for day-to-day operations. What happens when the third world country they off-shore to has the same thing happen? Unless GM forks over billions to "license" the now nationalized software, all of its plants and sales idle, putting hundreds of thousands out of work all at once. Same thing happens. Campaign contributions change hands and your children start coming home in body bags.
What scares me the most is that the off-shore companies themselves are going to force this to happen. Infokall, USTech, and the other large off-shore companies are built on a model of what amounts to slave labor. You are seeing articles in the business magazines about them complaining of talent sniping and a shortage of skilled developers willing to work for what they are willing to pay. Most of them are now opening offices in Korea and other countries which appear third world to Indian standards. These guys will pull out of their home countries overnight and open the door for some radical group to be backed by millions of now unemployed IT workers.
The move to Korea was really scary to hear about. U.S. troops have spilled blood there before.
Tyler: Roland, let's go back to your books. On your website, you state, "These books give IT people the information we actually need rather than the information the magazines say we need." What do magazines say IT people need that they don't, and why do the magazines have it wrong?
Roland: You have to understand how the "Industry Analyst" and trade magazine industry have operated for the past two decades to understand why neither are a good source of information. Both are funded by advertising dollars; both will deny it, but there it is. When a new product comes out and a vendor opens up its war chest, its first item of business is to become a paying subscriber to one or more of the "Industry Analyst" firms. This gets their product pitched to those in the IT industry subscribing to the service. It also gets Big-X consulting firms pitching the new product as well. Tons of articles appear in the weekly trade press stating how this new product is a Mega-Trend and the greatest thing to hit the industry since the semi-conductor.
This leads to knee-jerk decisions that launch countless "pilot projects" at various companies. These pilot projects all require some form of licensing for the product. The vendor then publishes this massive number of licenses being purchased (even if they are short term 120 day things) and suddenly it really looks like this is a train coming down the mountain at you. It's not. Until the new product replaces the actual core bread and butter systems at the company, it is nothing more than a flash in the pan. It takes a minimum of seven years to replace a core business system and have it settle in.
A core business system is defined as the complete flow: Order Entry, Customer Management, Inventory, Warehousing, Picking, Shipping, and Invoicing.
Let me put it to you another way. The language with the largest installed base in the world is COBOL. This is the language of many core business systems. There are millions of new lines of COBOL code written today and added to the billions of lines in production already. Exactly how many weekly or monthly IT magazines do you see writing articles about COBOL? None. It is a mature technology and doesn't have vast quantities of cash being dumped into its marketing.
Here is an interesting question for you to research on your own. Exactly how many college IT courses have COBOL as a mandatory course?
Tyler: Roland, I'm especially intrigued by your book "The Minimum You Need to Know About Logic to Work in IT." Your website suggests that logic isn't taught in college courses anymore, and consequently most IT people are unemployable. What do you see is the problem with IT college courses?
Roland: College courses are hamstrung by a lot of things, most of them fall into two categories: funding and tenure. I honestly thought that Y2K was going to fix college courses. There was evidence of it. Two years prior to Y2K hitting, a couple of forward thinking companies bought an IBM mainframe for a local junior college. They installed it and provided instructors. The governing body of the college was informed it would teach this course and actively recruit students for it. These companies knew that even graduating 50 students per term, they couldn't satisfy the need they were about to have inside of two years.
Tenure is a dangerous trap. It opens the door to some really lazy behavior. If you take a look at the college text market, the only books professors consider come completely packaged with test, scantron answer cards, overheads, and lecture notes. The instructor needs to add almost nothing to the course and in many cases doesn't.
Colleges don't have massive amounts of funding; even many of the private colleges only teach what they get for free when it comes to technology. Supporting a mainframe or midrange computer requires quite a bit of cash and special computer rooms. It is cheaper to scatter donated PC's around the campus and teach only what will run on them for free.
Colleges got trapped into trying to chase a market funded by a vendor war chest. When businesses said they needed IT professionals with WEB skills, colleges taught only the WEB skills. All of the other knowledge IT professionals were assumed to have didn't get taught. What you ended up with was someone who could design a really pretty WEB page, but couldn't communicate with the back end business systems or understand them. Why pay $65K/yr starting salary to a graduate like that when you can get the same unskilled person in a third world country for $10/day?
I have found very few colleges today that teach logic to IT people. The reason is that you can't make them understand how logic helps them if you aren't going to teach them the 3GL business system languages like COBOL, BASIC, C, etc. Logic is hard to understand in a point and click WEB world.
Tyler: Roland, when I introduced you, I mentioned that you are the president of Logikal Solutions, a business applications consulting firm specializing in VMS platforms. As a business consultant, if you were asked by a university that wanted to start an IT student program, to assist them, what would you do to make sure the students are prepared for the future?
Roland: They need to have the students spend their first three weeks (before committing to the program) studying the growth of off-shore companies, the labor rates being paid in those countries, and the unemployment rate among IT workers in the US. They need also to be informed of all the other career opportunities that are out there. They need to read the articles that have appeared in business and IT publications stating that IT workers are now "labor" and not knowledge workers as we were classified in the 70-80's.
Once the candidates have gone through that...assuming they start with 3-4000 for those first three weeks, they need to tell the one student that still wants to learn IT after all of that to go to another school.
Honestly, given the situation management has created in this country and globally, I cannot ethically recommend ANY college student to go into the field of IT. Until a tragedy of massive proportions happens, IT will not be a rewarding or well paying field. IT is currently not even respected by corporations anymore. MBA's sit through a one-day training course on how to create a contact manager using Microsoft Access, then get their certificate to manage IT projects. This is how we got where we are.
Personally, I do not think you will find an IT curriculum being offered at US colleges in fewer than five years. The last I read is that enrollment is down over 80% in IT programs nationwide. MBA's have themselves to thank. Some colleges have completely closed the curriculum and now only offer a few courses in WEB page design and Java coding for the WEB.
Tyler: What advice would you give today to students interested in pursuing an IT or programming career?
Roland: Right now, I would tell them not to pursue it. Become a water well driller or a diesel engine mechanic. IT is headed for a train wreck and we are less than five years away from it. The mad rush to treat IT workers like warehouse box stackers has lead to the beating down of IT salaries and massive amounts of fraud in the H1-B program. A small backlash against the off-shoring has already started with some high profile contract cancellations. The big hammer will fall when more H1-B workers get arrested by Homeland Security for acts of terrorism. After that happens, the H1-B visa will be abolished. Off-shoring companies will find themselves tightly restricted. You won't see thousands of IT workers slipping over here on vacation visas to work many months tax-free. IT workers will once again be respected as knowledge workers and salaries will reward those who know.
Tyler: Roland, what makes your books stand out and fulfill a need college courses have missed?
Roland: Logic is the fundamental tool of IT. If you do not understand logic, then you do not understand the fundamental principals behind IT. You didn't earn a degree; you were given one.
Tyler: Roland, I was surprised to learn your book "The Minimum You Need to Know to Be an OpenVMS Application Developer" is the first book in ten years on the subject. With the way technology is so rapidly changing, how is it possible ten years have elapsed without a book being written on the subject?
Roland: That's easy. HP is the third owner of OpenVMS. It started out with Digital Equipment Corporation who created an OS that was 30 years ahead of its time. Compaq then bought DEC, and being a PC company, had no idea what to do with a midrange system. Finally HP bought Compaq. HP has had a really sad excuse for a mid-range OS for many years. You might have heard of it: HP-UX. They sink vast amounts of money into marketing that lesser product. If that money were put into marketing OpenVMS, the HP-UX product would disappear inside of three years. HP is able to perform only maintenance on OpenVMS and have the OS add millions if not billions to its bottom line.
The installed base for OpenVMS is large. Companies that use it know what quality is. They also know the up-time for an OpenVMS cluster is measured in decades, not hours like it is for a PC network. Some of you may have read the article in "ComputerWorld" some time back. When the twin towers fell, the trading companies which were using clustered OpenVMS systems in multiple locations continued to trade until the end of the trading day. They had an outage of less than 15 minutes while the cluster verified the other nodes were not going to respond, then recovered their transactions and continued on. No other OS provides that level of "Survive the Fire" design.
Put yourself in the shoes of upper management at HP. You've sunk billions into this HP-UX thing over the years. OpenVMS has a large and loyal installed base despite every company that has tried to eliminate it over the years. Doing almost nothing for OpenVMS still has it adding millions if not billions to your annual bottom line. If you push OpenVMS, your flagship HP-UX will vanish from the market place. Do you tell the world you were wrong or do you continue sinking millions into HP-UX hoping against hope that it will one day catch up to OpenVMS?
Tyler: In "The Minimum You Need to Know About Java on OpenVMS," your first chapter is "Why Java?" Will you answer that question for us?
Roland: That question is best answered by reading the book.
Tyler: Roland, overall, what do you think makes your series of books stand out from all the other books on Java and programming?
Roland: I wasn't paid to write them. I wrote these books on my own time and published them with my own money. I wasn't paid by some publisher to crank out six books per year aimed at the least common denominator of the marketplace. This left me free to cover the topics I wanted and knew needed covering.
Tyler: Roland, what do you find most rewarding about programming and writing about our ever-changing technologies?
Roland: Technology really isn't "ever-changing." That's a phrase the trade press has been cramming down our throats for decades. Technology is forever rehashing old and sometimes bad ideas. The most rewarding part about writing is being able to point out just what idea is being rehashed this week by the trade press and "industry analysts."
Tyler: Roland, you have been involved with computers and programming for twenty years, back to when computers were just becoming common items in households. You have seen a lot of changes in that time. What have you found to be the biggest learning curve in keeping up with technology?
Roland: Convincing MBA's that what they are seeing in a 4-color glossy isn't new technology, it is a rehash of technology that either didn't survive or shouldn't be rehashed.
When you read through this series of books you will find a section where I cover how PC's rehashed mistakes mainframes and midrange computers made a decade before. You will also find a section talking about how all of these "new technologies" which let developers link directly to databases from WEB pages is a one way ticket to prison just waiting to be punched.
Tyler: Roland, you seem to have a bleak outlook for technology in the next few years. If you had a crystal ball, what would be your prediction for what technology and computers will be like in fifty more years?
Roland: Fifty is a really long number to look out. DEC had the best minds in the industry working for it and they only looked 30 years out. There are really three potential outcomes.
Outcome 1: Greed and corruption win. There are absolutely no IT jobs in the US, Western Europe, or England. Only a handful exist in Russia. All IT work is done by what was once third world nations. They bleed us dry. The former technology leaders now have a culture that exists of two classes, MBA's and those making less than $30K/yr no matter whether they build houses or work at 7/11. The domino effect caused by losing the IT workers caused a complete obliteration of the middle class by wiping out the industries which relied on them spending money (expensive homes, $70,000 SUV's, movie and music industry, etc.). It's the second dark ages.
Outcome 2: The SEC saves the world. During a brief respite between industry wide financial scandals the SEC stumbles into an accounting cover up of off-shore project failures by a blue chip company. They begin a very deep and public investigation. Heads of the company go to prison and the gory story of how papering over off-shore failures was common practice rattles the investing community. A cursory inspection of all publicly traded companies turns up that the practice was wide spread. In a massive plea bargain, all listed companies end their off-shore contracts within a month, then begin an examination of what systems they have still actually working. The mainframe and midrange systems still running their core business systems even after the company publicly declared they had converted everything to $800 pc's running Windows or Linux turn out to be the only system still running. A decade of purging happens during which, students are paid to go to college for core IT skills: Logic, 3GLs, and relational databases.
Outcome 3: Greed alone wins. The off-shore companies working in India faced with having to pay real wages and unionized programmers flash cut their operations over to Korea and other companies in a week's span of time. Millions of disgruntled IT workers take to the streets. Extremist groups move in and recruit them. These are educated people with a little bit of money, not the usual extremist fair. One or more large US companies finds all of their software nationalized by a new extremist government. We end up in a massive war with the outcome uncertain. Everything we want can be destroyed by a bombing raid or simply deleted by the current government of the country.
Tyler: Roland, would you tell our readers your web site and what further information they can find there about your books?
Roland: There are actually two sites. For information about the current books they can visit http://www.theminimumyouneedtoknow.com. For information about other books or my company in general they can visit http://www.logikalsolutions.com.
Tyler: Thank you, Roland, for joining me today. It has been a real education. I hope your books become popular and lead to wiser and better IT decisions and work.
No comments:
Post a Comment