When you hear hoofbeats do you think Zebras?
The last few days I've had a chance to help some customers with a few issues. As one of a group, I remember my tech support days and it troubles me that some engineers (both developers and test engineers) seem to be missing the basics.
When you hear hoof beats, think HORSES first, not zebras. All of these elaborate possibilities are considered rather than finding out by covering the basics. While I don't believe we should EVER be condescending to the end user, it's more likely we are missing some of the basic information than that some elaborate and exotic situation has caused the software to fail. When isolating a bug or troublshooting for a customer-look for the following first.
1. Scope
-How large is this problem? When did it begin? Has it always happened? Does it happen with other software? Can we even reproduce it? It is operating system or hardware specific?
2. Expectations
-Does the software even do this? Is this a workflow we expect to be supported? Many times as a customer or a tester we are expecting the software to do something it isn't even designed to do. One customer was using a server 3 versions old! That was never expected to work or tested with this version of the client software. Rather than all of these elaborate zebra stripes painted on the client, we needed an upgrade to the workhorse server badly. They can upgrade and migrate the data fairly painlessly if they know the server is the issue. GET the versions on everything. Investigate first, act later.
3. What is happening again?
-I will say as users go we have AWESOME users. They give us screenshots, exact errors, and often times even log files. Most software companies do not have users even half as educated as ours are. They certainly know their business and can show us things about our own software we didn't even know in some cases, however, they sometimes are blinded by their own focus and area of expertise, just like we are as testers. Take a step back from what you KNOW is happening and consider the bigger picture. JUST the facts. What are the experiencing overall. Take out the context and consider just the undesirable results. Not what they told you happened first. Not what they told you happened after. Just what is happening right now, this second. Sometimes they forget to tell you about the many beta versions they installed and didn't fully remove, or the custom plug-ins they have. There are times when just considering the current state you can more clearly isolate the issue than with all of the extra information. It's like one of the story problems in math when you need to strip out the irrelevant information in order to discover the answer.
4. What is our goal?
-Does this customer need to meet a deadline? Are we looking for a solution, or a clear problem? What do they need to accomplish? Can we do that a different way to help them first, and figure out what went wrong later? Do they need to get up and running, or can they help us understand what went wrong? Is what went wrong even important if we can fix it? Many times a problem will not be common and we can get them to a better state without even knowing the exact problem. Engineers hate that. Sometimes it will drive us crazy NOT KNOWING, but we need to give up and focus on the overall goal. Not in solving the puzzle, but in solving the problem at hand.
Are you testing for zebras as the herd of horses run by unchecked because you want zebras and they are more exciting? Are you looking for that bug that is a beautiful show pony when your workhorse is suffering for it?
This week I've had to ask myself these questions to get on the right track. What is important and exciting to us as engineers and technologists may not matter. Sometimes we have to get over ourselves to get to a solution and be the most useful.
One last thing, last week at PNSQC my co-worker and fellow conference attendee told me that someone at the conference said in surprise, "That's the redhead I saw you with?" in shocked tone after reading my paper. I asked her, why is that? Do I not look like I can write a paper? She just said, "Well, I think he was surprised because you are pretty." Of course I guess that was a shock to me because I don't see things that way. I am used to just being me. I am not always the same and while I enjoy makeup, glitter, and being girly, I don't see that as what defines me. I simply don't understand why anyone would want to be limited by that. One thing I've always loved about software is that what I look like isn't all that matters. I can be judged by my work and my mind, and my value isn't based just on what a look like. As a female this is one of the things I value most about being in software and being an American. I feel like I am not limited or forced in to one role because of my gender or what I look like.
I had someone ask me if I've ever tried modeling and I laughed. I said that I am interested in fashion, beauty, and art only as a hobby. I've never quite understood why some people place so much value on youth and beauty and I find it interesting. It isn't the top thing I value and it isn't what I consider the most important thing about me. I consider my appearance to be only a hobby and not very important for a few reasons. First, because it is something that doesn't last. I hope to age with character and grace. Secondly, because there is no winning. If you let what your appearance define you, you always lose. There is always someone younger, smarter, and more beautiful. Also, I know that despite the love of friends and family, you really need to be 16 years old and very emaciated for that job, and I'm nearly 2 decades past and on occasion I enjoy snacks.
The best thing about me has always been that I have heart and am sincere. I can't do better with intellect or beauty. I know that I am not one of the best minds in all of software. Many people are vastly more technical the intelligent than I will ever be. I know that I'm not one of the most beautiful people in all of the world and what I do have in terms of external natural decoration is not all my doing. I was born brightly colored. That isn't something to be praised for or that I pride myself on most. My ideas, my real accomplishments, the goals I've reached, that is what I'm proud of. Who I am as a person, and what is unique about me as an individual are what continue to matter over time, and that is what I appreciate and remember when someone else notices. While I can't deny that I need to hear that I am pretty and appreciated as much as the next girl, or that I have some brain activity going on, these are not the holy grail to me or "what I want to hear". I've been told that you should always tell a pretty girl that she is smart and you should always tell a smart girl that she is pretty, and you should always tell any man he is sexy because they all believe it no matter how far from the truth it may be. While in general that may be true and it may work most of the time, that isn't what I seek or value, and it isn't what the people I respect the most and aspire to be like want to hear. For me I just want to hear that you value my ideas, my work, and my contributions to the world. That you see and value me as a human. More of that interaction could really change the world, and not just software.
When you hear hoof beats, think HORSES first, not zebras. All of these elaborate possibilities are considered rather than finding out by covering the basics. While I don't believe we should EVER be condescending to the end user, it's more likely we are missing some of the basic information than that some elaborate and exotic situation has caused the software to fail. When isolating a bug or troublshooting for a customer-look for the following first.
1. Scope
-How large is this problem? When did it begin? Has it always happened? Does it happen with other software? Can we even reproduce it? It is operating system or hardware specific?
2. Expectations
-Does the software even do this? Is this a workflow we expect to be supported? Many times as a customer or a tester we are expecting the software to do something it isn't even designed to do. One customer was using a server 3 versions old! That was never expected to work or tested with this version of the client software. Rather than all of these elaborate zebra stripes painted on the client, we needed an upgrade to the workhorse server badly. They can upgrade and migrate the data fairly painlessly if they know the server is the issue. GET the versions on everything. Investigate first, act later.
3. What is happening again?
-I will say as users go we have AWESOME users. They give us screenshots, exact errors, and often times even log files. Most software companies do not have users even half as educated as ours are. They certainly know their business and can show us things about our own software we didn't even know in some cases, however, they sometimes are blinded by their own focus and area of expertise, just like we are as testers. Take a step back from what you KNOW is happening and consider the bigger picture. JUST the facts. What are the experiencing overall. Take out the context and consider just the undesirable results. Not what they told you happened first. Not what they told you happened after. Just what is happening right now, this second. Sometimes they forget to tell you about the many beta versions they installed and didn't fully remove, or the custom plug-ins they have. There are times when just considering the current state you can more clearly isolate the issue than with all of the extra information. It's like one of the story problems in math when you need to strip out the irrelevant information in order to discover the answer.
4. What is our goal?
-Does this customer need to meet a deadline? Are we looking for a solution, or a clear problem? What do they need to accomplish? Can we do that a different way to help them first, and figure out what went wrong later? Do they need to get up and running, or can they help us understand what went wrong? Is what went wrong even important if we can fix it? Many times a problem will not be common and we can get them to a better state without even knowing the exact problem. Engineers hate that. Sometimes it will drive us crazy NOT KNOWING, but we need to give up and focus on the overall goal. Not in solving the puzzle, but in solving the problem at hand.
Are you testing for zebras as the herd of horses run by unchecked because you want zebras and they are more exciting? Are you looking for that bug that is a beautiful show pony when your workhorse is suffering for it?
This week I've had to ask myself these questions to get on the right track. What is important and exciting to us as engineers and technologists may not matter. Sometimes we have to get over ourselves to get to a solution and be the most useful.
One last thing, last week at PNSQC my co-worker and fellow conference attendee told me that someone at the conference said in surprise, "That's the redhead I saw you with?" in shocked tone after reading my paper. I asked her, why is that? Do I not look like I can write a paper? She just said, "Well, I think he was surprised because you are pretty." Of course I guess that was a shock to me because I don't see things that way. I am used to just being me. I am not always the same and while I enjoy makeup, glitter, and being girly, I don't see that as what defines me. I simply don't understand why anyone would want to be limited by that. One thing I've always loved about software is that what I look like isn't all that matters. I can be judged by my work and my mind, and my value isn't based just on what a look like. As a female this is one of the things I value most about being in software and being an American. I feel like I am not limited or forced in to one role because of my gender or what I look like.
I had someone ask me if I've ever tried modeling and I laughed. I said that I am interested in fashion, beauty, and art only as a hobby. I've never quite understood why some people place so much value on youth and beauty and I find it interesting. It isn't the top thing I value and it isn't what I consider the most important thing about me. I consider my appearance to be only a hobby and not very important for a few reasons. First, because it is something that doesn't last. I hope to age with character and grace. Secondly, because there is no winning. If you let what your appearance define you, you always lose. There is always someone younger, smarter, and more beautiful. Also, I know that despite the love of friends and family, you really need to be 16 years old and very emaciated for that job, and I'm nearly 2 decades past and on occasion I enjoy snacks.
The best thing about me has always been that I have heart and am sincere. I can't do better with intellect or beauty. I know that I am not one of the best minds in all of software. Many people are vastly more technical the intelligent than I will ever be. I know that I'm not one of the most beautiful people in all of the world and what I do have in terms of external natural decoration is not all my doing. I was born brightly colored. That isn't something to be praised for or that I pride myself on most. My ideas, my real accomplishments, the goals I've reached, that is what I'm proud of. Who I am as a person, and what is unique about me as an individual are what continue to matter over time, and that is what I appreciate and remember when someone else notices. While I can't deny that I need to hear that I am pretty and appreciated as much as the next girl, or that I have some brain activity going on, these are not the holy grail to me or "what I want to hear". I've been told that you should always tell a pretty girl that she is smart and you should always tell a smart girl that she is pretty, and you should always tell any man he is sexy because they all believe it no matter how far from the truth it may be. While in general that may be true and it may work most of the time, that isn't what I seek or value, and it isn't what the people I respect the most and aspire to be like want to hear. For me I just want to hear that you value my ideas, my work, and my contributions to the world. That you see and value me as a human. More of that interaction could really change the world, and not just software.
Comments