While I Compile

… I compile my thoughts about programming

Consultants are advisors, not decision makers.

Overview

I was having lunch with my friend & colleague last week and we had a disagreement about whose decision it is to make a change when you see something wrong in the client’s software.

Mechanic analogy

My colleague used an analogy about a mechanic; it was a good one, so I’ll use it here.

Let’s say you bring your car in for a $30 oil change, and your mechanic notices a problem with the timing belt.  My colleague suggests, he doesn’t just ignore it, he tells you it’s an emergency must be changed immediately (the emphasis is my friends).

Well I agree the mechanic shouldn’t just ignore it, and I was relieved my friend didn’t suggest the mechanic should simply change the timing belt, driving the bill from $30 to $930, but I’m not sure if telling the car owner they must change the timing belt ‘now’ is appropriate either.

In my opinion the mechanic should tell the customer what he found, the risk in not fixing it, outline the options to fix it, and the cost & associated risk with each option[1].  Then make a recommendation.

It’s the customer’s decision, not the mechanics, and even if the customer makes a foolish decision, it’s his decision, not the mechanics.

Fortunately, all the mechanics I’ve dealt with seem to understand this.

As consultants we’re advisors, not the decision maker

In my opinion, our role as consultants / advisors includes the responsibility to inform the client of any problems or potential problems you’ve noticed, the risks in not fixing it, options to resolve the problem, along with the costs and risks associated with each.

Unethical behaviour

To me it seems unethical to go rogue, and just start making changes the client doesn’t know about, and didn’t approve.

It’s also unethical to purposely instil fear, uncertainty, and doubt when explaining the options to the client so they make the decision you want them to; regardless of your intentions.

The client owns your time

After all, the client does own the software, is responsible for its maintenance, and they do own your time to direct as necessary.

Wait, what?  Who owns your time?

Yeah, when you’re consulting, your client (or employer) owns your time.  … you sold it to them; remember?

Every consulting and/or employment agreement is different, but if you’re charging by the hour, it probably says something like this; they bought your focused efforts to solve their problems for a specific duration.

And how they spend your time (their resource) is up to them (within reason[2]).

So deciding feature X is more important than feature Y, or bug X is more important than bug Y, is their call, not yours.

An extreme example

But what if it’s a big important issue?  What if it’s a major security flaw in the banking software you work on.  The flaw is dangerous. The flaw could potentially put users at risk. The flaw could even push the company into bankruptcy.

But your manager doesn’t want to invest the time to fix it and his manager stands behind him.  You’ve exhausted all other paths of reason, to the point of being on the verge of getting fired.

You still don’t have the right to change it … not even on your own time, because the software is their asset, not yours.

Your decision

Now I’m not saying you have no say, you can refuse.  Bottom line: if they refuse to make a change that you feel is necessary and/or potentially dangerous, you can give notice and quit.

Exceptions

Like everything, there are exceptions to this general rule.  Here are 2:

The first is if you are working on one task, notice another problem, and can fix it without seriously adjusting your time budget on the original task.  An example of this might be noticing and fixing a 1-off error in a loop.  I once had the responsibility of making manual year end changes to customer databases, with an estimated time of 3 days for each.  It took me 4 iterations working smart within the allotted time to write and test a utility which dropped that task to a 1 hour job[3].  The utility was separate, so I didn’t change the base product, and it cost my employer nothing.  So it is possible to make an impact working like this.

The second exception is you have built up a lot of trust with the client, and are completely unsupervised, making whatever changes deemed necessary to accomplish the client’s goals.  I’ve been in this situation a few times, and can honestly say, it’s a rare situation to be in.  It’s reserved for the situation where the client trusts your character and your responsibilities are so mysterious that they cannot make a decision, so they leave it up to you.[4]

In Summary

So basically, in my opinion, as consultants it’s not within our rights to make the decision to change or coerce the client into allowing you to change something.  It’s our obligation to inform them of the problem, outline their options along with the risks and cost of each, and make recommendations.

… that’s it.


[1] Yeah, there aren’t too many options to discuss for changing a timing belt, but we’re discussing software remember.

[2] There are limits of course, everybody will have their own personal & ethical limitation, and many (not enough though) will have professional limitations of what they’re willing to do.  Professional limitations might include, I was hired as a programmer, not a janitor, or worse yet, and Access developer.  ;-)h

[3] I could’ve dropped it to 5 minutes, but 1 hour is where the 80/20 principle told me to stop.  Somebody did this after I left, and it took significantly longer.

[4] Neither my colleague or I are currently in this situation.

February 2, 2011 Posted by | Consulting | , , | 2 Comments

When to start looking for another job

Last week I listened to Seth Godin’s audiobook Linchpin. It really resonated with me and I had a few significant insights.

One of the interesting ideas he promoted was the idea of ‘gifts’. A gift as he explains it, is any additional work above and beyond what is required as part of the ‘transaction’. The transaction is fulfilling your end of the bargain for their end of the bargain. Seth Godin explains a transaction as

If I sell you something, we exchange items of value. You give me money, I give you stuff, or a service. The deal is done. We’re even. Even steven, in fact.

And he explains a gift as

If I give you something, or way more than you paid for, an imbalance is created.

Lets say a client is having an issue and after some digging, you have an insight where a slight change not only resolves the current problem, but prevents a similar problem from occurring throughout the entire application. The client did not offer to pay for it, and you can’t charge them, as a matter of fact, if you are a consultant, it will reduce future billable work from the client to fix the future problem. The insight and change is a gift.*

Seth continues with regards to the gift:

That imbalance must be resolved.

So how is this imbalance resolved? … Appreciation

Yep … that’s it.

If you have a particularly astute client/employer, you may receive additional work and referrals as a consultant, or a bonus, raise, and/or promotion as an employee, but these are peripheral. Appreciation is the critical element for the recipient to experience**. If your gifts are not appreciated, the client/employer does not value what you have to offer and both of you should seek out more compatible relationships.

So why give a gift? Personally, I give gifts because I want to create the best software I am capable of creating. Creating beautiful software, not just meeting requirements, is the very nature of craftsmanship. Beautiful software is a gift. I can’t write software without gifts. I am actually repulsed by the thought of merely delivered what was asked for since the requirements are always missing something. Yes I’m repulsed. It reminds me of those Mad magazine comics; ‘If kids designed their own xmas toys’. With few exceptions, the results would be horrendous!

Unfortunately, I’ve noticed a downward cycle with the reception of my software gifts. I’ve noticed when starting a new position, gifts are recognized and appreciated. You are ‘the man’ (or woman) and everybody is ecstatic with every gift. However the appreciation always seems to dissipate. Maybe it’s a ‘familiarity breeds contempt’ type of thing, or perhaps it’s just me (but I don’t think it is).

I believe there are 5 phases of perception regarding the receipt of your gifts:

  1. Worshiped – You are new with the client/employer and your gifts are truly unexpected, recognized as such, and are appreciated. You get thanked a lot by every recipient.
  2. Valued – You’ve been here for a while, and although these above and beyond tasks are appreciated, they’re not exactly unexpected anymore. You don’t get thanked much, but they realize you are a valued service provider.
  3. Unappreciated – Your gifts are expected and/or unrecognized, but unappreciated any way you slice it.
  4. Tolerated – Your gifts are viewed as a time consuming waste of effort, but tolerated. You continue to provide them out of a desire to do a good job.
  5. Rejected – Your gifts are rejected and no longer tolerated. Every suggestion of doing something with additional benefit is rejected.

In my opinion, the transition from Worshiped to Valued is normal, expected, and even desired. People can’t run around thanking you for the rest of your life, nor would you want them to. It might get a little creepy. 😉

The transition from Valued to Unappreciated is your cue to leave. This is a downward slide, and it’s unlikely that things are ever going to move back up to Valued***. You are in a great position to find other work, you have plenty of time to find the ideal next job or project, and you are leaving on a high note with a favorable memory still in their minds. However, you do need to be objective in your observation, your gifts may still be recognized and appreciated, but the feedback you are receiving is based on another pressing issue at the company and is misleading.

If you’ve moved to Tolerated or Rejected; you have completely missed your cue to leave and there is an obviously serious disconnect between what you are offering as a gift and what management perceives as valued. Regardless of the reason, both you and the client/employer might need to seek more compatible relationships***.

So what if the problem is not in the ‘perception’, but you have actually become complacent and are no longer delivering the gifts? If this is the case, you had better get back on track, because these gifts are your value added proposition and the only thing separating you from the lowest cost outsourcing alternative.

EDIT (03/08/2010): Ben Alabaster added a great comment about how an IT department being viewed as a Cost Center or Profit Center will impact on how your gifts are perceived.

*This is not a gift if the change took a significantly larger amount of time which the client did not agree to.
** The appreciation does not have to be communicated, but it must be felt by the recipient.
*** It has occurred to me that gifts could be ‘adjusted’ to more closely align with what management values. However, I think this idea is flawed since your ‘gift’ is your best ideas at improving the software, where as management is mostly interested in features, which is completely different.

Copyright © John MacIntyre 2010, All rights reserved

March 8, 2010 Posted by | Career, Consulting | , | 7 Comments

2 Rules for a Happy Client while Billing Hourly

Some of my projects are a fixed rate and some are hourly. If the scope of the project is small enough to accurately estimate, risks are minimal, and nailing down requirements is practical, I may propose a fixed price. Working on fixed price projects are a piece of cake, I go home, work on it, provide periodic updates via email, phone, or in person. When I finish it I install it and give them a CD with all the binaries, documentation, and source code.

However, I’ve found working on an hourly rate to be a little different than doing all the above while just counting my hours. There is a lot of distrust with hourly professionals which is only compounded with a mysterious process like software development. After all how do they know you aren’t home watching soap operas, working on your own projects, the project of another client, or even sitting on the beach? How can they tell they aren’t being over charged? The problem with trust, is even if you are honest, a false perception can still destroy the relationship.

I know this too well. Once in the late 1990s I was working from home, charging hourly, and I got a call from one of the newer partners of my biggest client (only client at that point). While talking I rebooted to clean out my system* and when the system came back up it played the 20th Century Fox theme, a configuration change I thought was very cool, and he said ‘Are you watching TV?’ I said ‘No. Why?’ … Seriously, I didn’t even making the connection. Two days later I was called in for an unscheduled meeting where my invoices were questioned, I heard a lot of “we’re not accusing your of inflating your hours…”, and the next day, feeling insulted and unappreciated, I resigned. I didn’t make the obvious connection until much later. It doesn’t matter that I was under charging both in terms of my rate and what I charged for, and would never inflate my hours! It doesn’t matter, because the perception was corrupted. It was a tragic misunderstanding since I loved working there and provided a competent, yet naively discounted service.

That’s when I came up with rule # 1
1. Work at the client site when charging hourly.

While following this rule gave my clients assurance that I was actually working, it still left them in the dark as to my effectiveness and what exactly I was doing. I soon realized that working on site isn’t enough. I need to communicate my challenges and accomplishments more effectively as well.

So I came up with rule # 2
2. Provide detailed invoices.

Until this point, my invoices had the typical, single ‘All services rendered’ line item. I changed it to include every single thing I did. I’m not kidding. Here is a sample from an invoice I wrote a couple years ago, after establishing this rule:


App F – Investigate & Fix bad XXXXX data

App A – Altered data tables to use the datetime data type for the create_dt & update_dt fields. This is needed to have a granular time for list updates. Communication /w Coder X and answering his questions. Conversation with Coder X about Stored Procedures. Discussions with Coder X about the stored procedures. Generating Business objects.

Feature X – Add Save & Save to Profile to the IFeatureX specification. Alter the Feature X specification document to include autosuggest searching ability in the IFeatureX interface. Reviewed Coder Xs progress.

Planning – Discussions with Manger Y about the purchase of App R, project status, and got permission to take the API documentation for App R off-site for review.

Website – Added new user.


Notice how it’s written for the target audience (a non-programmer, IT manager in this instance)? Notice all the detail? Notice I explained why I was changing the data types? Notice the descriptive verbiage (investigated, fixed, altered, reviewed, etc…)? Notice how even something as basic as adding a new user to the website, a 5 minute task is included? Notice I confirmed my permission to take confidential documentation off premises?

There is some terminology which might be unfamiliar to my client, but these terms were introduced to the client before hand, so they understood every thing said on that invoice when it showed up in their inbox.

All this detail may seem like overkill, but it gives the client a level of transparency into the mystical world of software development. It provides a record of activities, allowing the client to feel in control, knowing the priority decisions they made are being acted upon as agreed. It reveals a shadow of tangible evidence on an outwardly invisible service. But most of all, it gives the client comfort and I believe raises the level of trust.

I should also point out, the invoice sample above is not based on a 50-60hr work week, but from a week where I put in 14 billable hours. A 50-60hr work week would usually be 500 words or more.

… hey! Don’t freak out, you can write a detailed invoice without spending your weekend on the first draft. Here’s a brief list of things you can do to reduce the time you spend writing monster invoices:

  1. Maintain a detailed time log filled out as you do things. This takes discipline, but almost no time.
  2. Make detailed source control comments and do a report at the end of the time period.
  3. Make detailed bug tracker notes and list all the bugs/tasks/tickets you worked on during the time period. You can just list the bug ids, but I usually write something like ‘Investigated and resolved ticket # N – User receives 404 error when clicking customer link’.

The key to any of these is to write with your client as your intended audience, this way you just cut and paste into your invoice.

More than a decade after reluctantly getting into consulting, I’ve realized that Perceived value is directly related to the quantity and quality of communication with the client. An invoice can be a key communication device with your client and a powerful marketing tool if you make the effort.

*It was Windows 95 after all. At that point I was rebooting 10 times a day at least.

Copyright © John MacIntyre 2009, All rights reserved
WARNING – All source code is written to demonstrate the current concept. It may be unsafe and not exactly optimal.

July 9, 2009 Posted by | Consulting, Non-Programming | , | 1 Comment

11 Personal Programming Assumptions That Were Incorrect

Today I got side tracked and spent an unreasonable amount of time on StackOverflow.com. One of the questions I was looking at was What is your longest-held programming assumption that turned out to be incorrect?

Many of the answers immediately resonated with me, like Instantsoup’s answer That people knew what they wanted and JohnFx’s awesome answer about comparing his knowledge to the collective knowledge of all other programmers. Other answers reflected a poor initial understanding of the language or technology, many of these I was fortunate enough to not relate to.

As you can imagine, I immediately started coming up with my own answers, so I continued reading to make sure they weren’t already there. But as I read, I came up with more initial assumptions which proved to be false. I thought I’d pick out the best, and answer with that one, but realized I had a whole blog post!

So without further ado; here is my list of assumptions about programming and the industry which proved to be incorrect:

  1. The customer and user are the same person.
    In consulting with custom in house software this is (almost?) never the case. The customer has their eye on the budget and many won’t give 2 cents to make something easier for the user.
  2. You isolate and kill all bugs without exception
    Truth is; there will always be bugs, and what gets fixed is an ROI decision.
  3. Writing beautiful software as an act of craftsmanship
    It’s taken a LONG time for me to get it through my head that it’s just another financial investment, where you want to get as much as possible while giving as little as possible. However, although I know it intellectually, I still don’t ‘get it’.
  4. Working 24/7 would be rewarded
    Truth is, working all the time will burn you out, lower your productivity, and cause you to make stupid mistakes which reflect on your professional abilities.
  5. Vendors can be believed
    In my experience, don’t believe the product will integrate as seamlessly as they say without getting their API documentation and building the whole integration mentally first.
  6. You are not actually working from the monitor.
    With today’s GUI platforms, this is largely abstracted out, so you can actually program believing that a textbox is a physical entity. But when you are managing your own graphics, that textbox isn’t going to capture your keyboard presses and display them, unless you provide the illusion. It’s actually up to you to capture the keyboard input, and adjust the memory buffer containing your screen image with the newly typed letter in the control. This was a real mind bender for me, when I first got into this.
  7. That I wasn’t a very good programmer
    Like JohnFx’s answer, I was unsure about my abilities. I don’t have a C.S. degree, and was overwhelmed by the depth and breadth of the knowledge out there. But the more people I meet, the more I realized, I might actually be pretty good. Don’t get me wrong, under malicious attack, there are many who could expose and exploit the holes in my knowledge, but in my general area (business software), I’m finding that I’m not too bad.
  8. You need to / should grok a language or tool before you even start.
    Groking is a powerful way to start with a new language or tool. But in reality, especially with consulting, there isn’t always time. If you can hack something together in a completely unfamiliar language/tool/framework/paradigm/etc in 5 days, and it would take somebody who knows it cold 1 day, but it would take you a month to grok it … you hack. You may not want to, but from a business perspective … that’s the correct answer.
  9. You don’t say you know something unless you’ve grok’d it.
    Well, you don’t need to have too many conversations with too many programmers to know this idea is not wide spread.
  10. Other programmers saw beauty in their work.
    Personally, a well written piece of software is similar to a piece of art. It’s my form of artistic impression. It’s my baby. However, there are many people in our industry who do this 9-5, and appear to get no more enjoyment out of it, than I got out of a factory job I had in my teens.
  11. The best programmer is recognized
    In retrospect this is almost laughable, at least in consulting! If programmer A writes a wiz bang sub system, takes 2 weeks, it’s flexible to extend, and the core system never requires maintenance. Programmer A would not be held nearly in as high regard as programmer B who takes 1 week, hacks together a piece of garbage which is a nightmare to extend, and is maintained with multiple bugs for it’s life time. This is unfortunate, but in my experience, only a very capable manager would know the difference.

Copyright © John MacIntyre 2009, All rights reserved
WARNING – All source code is written to demonstrate the current concept. It may be unsafe and not exactly optimal.

June 4, 2009 Posted by | Programming | , , | 12 Comments

AboutMe.xml

I have a pet peeve and like most pet peeves it’s an irrelevant petty little annoyance, not quite a huge, humanity, oppressing problem.

My pet peeve is filling out the same information; name, address, city, etc… on paper forms. All that standard information at every doctor’s office, school, activity registration form for my kids, etc… I mean why do I need to keep writing this stuff? And why does somebody else have to take the time to retype it into their system?

Really! In all seriousness … what a waste of time! 5 minutes I’ll never get back, every time I start a new relationship with any organization.

But wait … I have a vision! Not a big glorious, save humanity vision, it’s more of a save each person 5 minutes of writers cramp, kind of vision. Yes! That kind of glorious vision!

I was originally inspired with this in the mid 1990’s. It started out as a question; why can’t doctor’s receptionist retrieve this information from the province when they scan my health card. But since the likelihood of getting the government to add an API for this is slim, it was reduced to something simpler. Like; Why can’t I hand the receptionist at my new doctor a diskette with an ‘aboutme.txt’ file on it, where she can load it into her PC, and give me my diskette back? This would free me up to spend an extra 3-5 minutes browsing the 4 year old magazines during my 76 minute wait to see the doctor.

Over the years, this vision has transformed from an aboutme.txt file on a 3.5” diskette to an aboutme.ini file on a diskette to an aboutme.ini file on a website to an aboutme.html file on website to an aboutme.xml file on a website to an aboutme.xml file on a USB memory stick. I’m not even going to go into ideas I had for RFID, bar coding, or carrying around printed labels in my wallet.

I’ll agree; this isn’t a big problem, but it’s an irritating little annoyance which can be easily overcome with a very simple programming solution. Surely, this would become a reality. Surely, this simple idea would be recognized by others, and implemented.

But alas, the obvious was never realized and because it would be impractical for any organization to expect you to have this aboutme.* file in your back pocket when nobody else had one or was asking for it. It’s the typical chicken / egg scenario; you need one to start the other.

But now I’m inspired again … by Open ID, or possibly another similar centralized authentication mechanism.

When I log into a new site via MyOpenID, I can chose the persona I want revealed to the site I’m logging into for the first time. One of these personas could easily contain standard address information like that required in the types of situations listed previously.

As Open ID reaches critical mass, with more people understanding and adopting it, providing and/or recommending software functionality to accept basic information via an Open ID login will become more realistic.

It’s easy enough to imagine a plausible working process, so I won’t bore you with that. However, there would be serious security concerns regarding logging into a critical authentication mechanism like Open ID from a shared kiosk, so the user would want to log in via their personal cell phone (or laptop or PC or …). And mainstream user adoption has a long way to go before something like this would even be offered, not because of technology, but due to slowly shifting paradigms.

There are obstacles to overcome before this could ever become a reality, but with centralized authentication schemes like Open ID, expecting most people to have an electronic copy of their basic information available will eventually be reasonable, and generic business software applications will start consuming that information.

And one day, hopefully before I die, I won’t have to fill out another one of those stupid forms.

Copyright © John MacIntyre 2009, All rights reserved

WARNING – All source code is written to demonstrate the current concept. It may be unsafe and not exactly optimal.

May 11, 2009 Posted by | Non-Programming | , , | 3 Comments

   

%d bloggers like this: