The Whiteboard

4:34 AM

Programmers forum might be a better place to ask. — Santhos 56 secs ago

5:09 AM

@gnot It seems to me like a design/architecture thing, definitely a better fit for programmers than for SO imho. — Santhos 19 secs ago

Duga

5:44 AM

@Santhos Please stop using Programmers.SE as your toilet bowl — gnat 56 secs ago

Duga

7:45 AM

Voting to close because this question is not about a specific programming problem. The "programmers" sister site may be more suitable — Joni 51 secs ago

Kit Z. Fox

11:53 AM

Did you hear yet, @Thomas?

Thomas Owens

@KitZ.Fox ?

I heard that the bird is the word.

Kit Z. Fox

About the thing. Now I can't remember if I'm supposed to talk about it. It's like Fight Club all over again.

Thomas Owens

If it's that thing, not yet.

Kit Z. Fox

And also maybe it wasn't you.

OK.

Thomas Owens

Soon.

Kit Z. Fox

11:55 AM

Happy Coffee Day.

I was in your neighborhood recently and did not get a beer with you like I said I was going to.

I did peek through your windows though. I like your couch.

Thomas Owens

My windows are on the third floor. How tall are you?

And they are tall (10ft) ceilings on each floor. So you must be like...35 ft tall?

Kit Z. Fox

Um. I hung down from the roof.

Or ... I might be lying.

My scrum team has dissolved into a mass of confusion.

So I thought I would stop in and ~~complain~~ ask for advice.

Thomas Owens

So they aren't people anymore? I think the first thing to do is to clean up this amorphous mass of confusion that is likely on the floor and try to get a new team in place.

Kit Z. Fox

But it's too early for that, so I'm drinking coffee and subtitling a video and trying not to be angry with myself that I'm procrastinating.

@ThomasOwens haha. We have two new people.

Thomas Owens

Ah, fresh minds to corrupt and twist.

Kit Z. Fox

12:04 PM

For a total of eight at the standups. Sometimes nine.

And the director and one of the business stakeholders keeps telling me to add more people to the backlog grooming meetings and the reviews.

So I've got somewhere around 20 people on those meetings.

Thomas Owens

9 is the high end for a Scrum Team. That seems OK at standups, although they may tend to be a tad bit longer than 15 minutes.

Kit Z. Fox

Yeah, it would work if everyone knew what they were doing, but they don't.

Thomas Owens

Although it's "not Scrum", I'd probably be working toward a 20 minute timebox with 9 people.

Kit Z. Fox

The scrum master apologized to everyone yesterday for having to remind them to actually talk to each other outside of the daily standup.

Thomas Owens

It'll take some time to get there as people learn.

Kit Z. Fox

12:07 PM

Yeah, but no one is teaching.

Thomas Owens

Then what is the Scrum Master doing?

Kit Z. Fox

He doesn't think it's his job.

Thomas Owens

That's one of the biggest things for that role to do.

Kit Z. Fox

But the guy who pushed for Agile, the scrum master's boss, doesn't feel that the team needs any training because "all the guys are basically familiar with it because they've done it before in other jobs".

Except that it is obvious to me that we all have different expectations about the roles.

And today I am faced with the question: What do I do when the release for Sunday has not been tested and is not ready for UAT?

Thomas Owens

That attitude may be fine at some points, but it doesn't last as you try to mature.

@KitZ.Fox Is anything tested and ready for UAT?

Kit Z. Fox

12:11 PM

And what I'm getting is that the dev team is saying "yes, it's ready for UAT" and SQA saying "nobody gave us these things to test".

@ThomasOwens Two out of 8 are tested and not passed.

Thomas Owens

@KitZ.Fox So nothing is passing tests?

ratchet freak

@KitZ.Fox then someone needs to say, pass everything ready for UAT to SQA

Kit Z. Fox

I have no effing idea at this point.

I don't even know why we have standups.

And I feel totally powerless to do anything except watch.

Thomas Owens

At this point, you can cancel this Sprint. Use today and tomorrow to clean things up, though. Go into your Retrospective and fix things. Or, you can work on getting a couple of things passing tests and into UAT and deliver something. It may only be one or two things, but 1 is better than 0.

Kit Z. Fox

We had to rollback the entire last release because one of the major pieces wasn't tested.

But how can I cancel the sprint? I don't have that authority.

And I'm the only one who seems to think there's an issue, despite what happened last time.

Thomas Owens

12:13 PM

I think the Product Owner does that...

ratchet freak

start a rant on the next standup

Kit Z. Fox

I don't think I can be constructive at this point. I want to pitch my desk over and scream at people.

Kit Z. Fox

12:25 PM

I told business that we were going to delayed the release of a form and had a meeting about it yesterday where she basically told me it had to go out and the guy in charge had nothing to say about it.

I'm not getting any backing.

I don't even know what my job is anymore.

maple_shaft

1:09 PM

@ThomasOwens I hope that you aren't sore about your opinion not being well received in Meta

For what it is worth, I somewhat agree with you that the site name Software Development might be more responsible. I downvoted you merely because I think your answer was too ambitious

Thomas Owens

I'm actually refining it now.

I was just about to post when I got your ping.

maple_shaft

oh

Thomas Owens

I stole ideas from you and @RobertHarvey.

Ixrec

I'm still wondering what exactly caused that reaction, my best guess is the perception that you wanted to remove "80%" of what's on-topic here

Thomas Owens

0

A: New Site Name and Scope Proposals

Site Name I've started to come around to the name "Software Engineering". Although I still have some lingering objections regarding the overuse of the term "engineering" in software development, I think that the benefits would outweigh my personal objections. It does bring to mind a professiona...

maple_shaft

1:12 PM

I think we all have our pet topics that we like to push

Thomas Owens

WOAH! Law is Graduating!

17

Q: Congrats, Law SE will be graduating soon!

Good news, everyone! Or, should I say, "good news: everyone"? In the year that Law Stack Exchange has been open, we on the Community team have been really impressed by how well things have gone here, and credit for that goes entirely to the amazing effort and knowledge all of you—and especially y...

2

maple_shaft

for me, I am a strong believer that the term Engineering is overused and abused... however I STRONGLY feel that as a profession, we need to mature beyond just "Software Development" and rally around a strong Engineering professional organization

Ixrec

O_O

Kit Z. Fox

You mean the name still hasn't changed?

Thomas Owens

Why is Open Source not graduating, too? Anyway, can we just kick all of our licensing questions to someone else now? I think that's the only thing I'd like to descope. It's just hard to explain what licensing question is on or off topic.

Ixrec

1:14 PM

we also get asked more off-topic licensing questions than on-topic

TZHX

I rarely see good questions around licensing. the two most recent examples I saw seemed to boil down to "I know it's against the spirit, but am I technically breaking the law doing [convoluted thing]?"

Ixrec

the ones people really want to know are all the ones that haven't been tested in court

there was a phase where we got a reasonable number of answerable ones and I answered them with quotes from the GNU FAQ, but these days I just don't see any I can do that on, and OpenSource.SE kicks our *** at those questions anyway

Thomas Owens

Yep.

ratchet freak

@ThomasOwens there are other license than just the open source ones

Ixrec

those are the only ones we can ever seem to answer

ratchet freak

1:16 PM

granted those could be answered in law as well

Thomas Owens

@ratchetfreak And if you dig on Meta, I think we decided that things that aren't commonly used open source licenses aren't valid. If you get a license when you buy software, we don't deal with it.

maple_shaft

@ThomasOwens I love your new answer btw, upvoted

Thomas Owens

My answer = you + @RobertHarvey + bits and pieces from other comments and answers. But it's mostly merging my thoughts with yours and his.

maple_shaft

I still struggle to say we should remove algorithms and data structures though

i think NOT spelling it out might alienate a bunch of people that would otherwise ask a great question around these topics

People that might be amateur enthusiasts or students that are a little intimidated by the high level nature of the stated scope

ratchet freak

@maple_shaft which can fit on compSci.SE

Ixrec

1:19 PM

wait, was anyone in favor of removing algorithms/data structures?

Thomas Owens

@Ixrec From the list, but not the site.

Ixrec

yeah that's what I thought

maple_shaft

@Ixrec No, everybody still wants that to be on topic.

@ratchetfreak I understand that it overlaps with Comp Sci a bit.

Thomas Owens

It's just a different perspective than CompSci. I'd like to communicate that somehow.

Ixrec

there are a lot of alg/data questions that aren't purely mathematical, or that the OP has no idea how to phrase mathematically

no need to rule out anything based on overlap

maple_shaft

1:21 PM

@Ixrec THIS.

Ixrec

but those are usually part of what's otherwise some sort of "design" question

Thomas Owens

Something else I'd like to do is somehow identify these other related sites that we aren't linking to as a place to go for off-topic questions. Computer Science and Project Management are probably the two biggest ones.

Ixrec

so I'm cool with potentially not giving them a separate bullet point if we want to be really really compact

maple_shaft

Some of us don't have formal CS training. We are either self taught, or taught in a software engineering curriculum where it may be harder for us to communicate in mathematics and theory

Ixrec

I have a little formal CS training, but I'm nowhere near the level of the guys on CS.SE

Thomas Owens

1:22 PM

@Ixrec That's exactly my thinking for why I proposed dropping it from the list. But not the site's scope. They are design questions

Ixrec

Cook-Levin Theorem is the hardest thing I understand

@ThomasOwens yep, I'm agreeing with you

Thomas Owens

@Ixrec Tell me about it. I don't even know what that is.

Ixrec

@ThomasOwens that's the one that proves NP-completeness is a thing

by proving SAT-3 is NP-complete

Thomas Owens

I failed a Google interview hard because they wanted me to start writing my own data structures and stuff. I'm like "...I do software engineering? Where are the requirements, design, test questions?"

Ixrec

you mean you never implemented a linked list in C with pointers for no reason?

maple_shaft

1:23 PM

@Ixrec Cook-Levin theorem, thats where you prove that by adding yeast to bread, when you cook it in an oven it levin's right?

Ixrec

@maple_shaft it can probably be used to prove that a simulation of what you just described is NP-complete

maple_shaft

or the inverse rather

@ThomasOwens Yeah their interview is very CS academic slanted. I crammed for a few months on stuff like that before my Google interview. Funny enough the only part of the interview I bombed was the Software Engineering part

Thomas Owens

I didn't even have a Software Engineering part, really.

maple_shaft

Really? One of the 5 or 6 interview sessions I had was focused on high level design on a whiteboard. The problem was so complex that I embarassingly floundered

They were like, "Design Google Suggest... Go."

Thomas Owens

Well, it started as a design question. And then turned into "so, what algorithm would you use to implement that?"

maple_shaft

1:28 PM

I started trying to gather requirements from the interviewer and literally couldn't even wrap my mind around the numbers he was throwing out there

Ixrec

@maple_shaft would I fail just for having never heard of Google Suggest?

maple_shaft

@Ixrec No, he would have explained it to you

Its when you type in the Search field and it offers suggestions in an auto complete

Thomas Owens

That has a name?

maple_shaft

I think its called Suggest

ratchet freak

fuzzy search + typo correction

and some popularity stats

Ixrec

1:29 PM

I thought it was called "autocomplete"

maple_shaft

regardless I was trying to design an in memory distributed cache that was horizontally scalable and duplicated indexes

then he started challenging me on just how large the indexes were... and when I realized he was right my mind turned to spaghetti

and I started trying to write on the whiteboard and it was all moms spaghetti

Thomas Owens

I don't think anyone is used to thinking at Google Scale.

maple_shaft

Yeah I honestly dont know how anybody passes a Google interview

ratchet freak

maybe they don't expect people to pass them

maple_shaft

Search has the top minds of the world working on it, and you give me that crap in an interview question for 45 minutes?? F--- you

Ixrec

1:34 PM

@ratchetfreak thecodelesscode.com/case/83

2

> Therein lies its value ... For we are all of us doomed in this profession ... how will he choose to fail? By cowardly surrender? By costly victory? By erroneous compromise?

maple_shaft

I cowardly surrendered, but when I went to sign the surrender, spaghetti spilled from my sleeves onto the document

Kit Z. Fox

I choose to fail spectacularly and by a means of my own choosing.

Ixrec

this is where Aaron tells us the magic of C3 in Python lets him give the soldiers flaming arrows without modifying any of the base classes

Thomas Owens

Oh, speaking of topicality...tools. Are we the only site that doesn't allow tool questions that are about things in the on-topic definition? Engineering allows AutoCAD questions, Cross Validated allows Minitab questions, and PM allows JIRA and Microsoft Project questions. Saying "no tool questions" is easy, but I'm not sure I want to go to Stack Overflow and ask about using an ALM tool or DOORS. I don't want to replace a vendor's support channels, but I think "no tool questions" is strong.

Ixrec

what kind of tool questions would we want though?

we have some stuff about git branching strategies I guess

Thomas Owens

1:45 PM

Maybe most of the questions I'm thinking of should go to a vendor or consultant and not the Internet.

Ixrec

yeah, I can't think of any tools that are unique to software design/engineering yet not something SO would already be far better at

in many ways even the core of questions we do want is the "soft part" of what SO covers

ratchet freak

source control?

Ixrec

5 mins ago, by Ixrec

we have some stuff about git branching strategies I guess

ratchet freak

git is far from the only source control out there

maple_shaft

Replace git with 'decentralized version control system' and you have a good question

Tool questions are rife with problems

Ixrec

1:54 PM

yeah, I think a dvcs question that is mostly not specific to a single dvcs tool is a good fit here

Thomas Owens

I suppose "no tool questions" is OK.

maple_shaft

Certain frameworks like .NET and disciplines like Engineering don't really have many choices for what it is you are trying to accomplish

Thomas Owens

You can ask about your process, and if you have an existing tool, you can mention that you have something.

So than answers can be framed in the terminology of your tool. But if you can't figure out how to make your tool achieve what we tell you, you should call the vendor or a consultant.

maple_shaft

I want to draft a design, AutoCAD. I want a rich presentation framework in a desktop app, WPF

I want dependency injection in Java...

421 unique answers

there are just too many choices out there for some frameworks

it becomes an opinon war

Can I go to Unix.SE and ask, "What distro should I install?

Thomas Owens

But there's not just AutoCAD. There's Creo.

And SolidWorks.

OK. I think my Meta answer is done.

maple_shaft

2:00 PM

If [industry accepted and popular framework] <= num[maple_shafts testicles] Then [tooling question] == Status.ACCEPTABLE

Kit Z. Fox

Hmm.

That raises an interesting question.

maple_shaft

btw... I had a terrible bicycle accident when I was 13

later

ratchet freak

@KitZ.Fox do we even know if maple_shaft is male?

Thomas Owens

Yes.

Kit Z. Fox

2:16 PM

I don't have positive evidence for that.

@maple_shaft Do you identify as male or prefer not to say?

Hmm, now wait.

That's another sticky wicket.

Thomas Owens

2:29 PM

Am I the only one scared by the possible outcomes of Oracle/Google?

Kit Z. Fox

It will be Magic 8-ball then, right?

Thomas Owens

I don't know. But I don't like how it's in the hands of a jury that doesn't even get how software works and the implications of their decision.

Duga

This question is off-topic for Stack Overflow because it is about design practices and is inherently subjective. You might consider asking about considerations when making these types of decisions on Programmers.SE. — Cody Gray 1 min ago

ratchet freak

2:48 PM

@ThomasOwens well the javadoc license is pretty clear...

0

A: Oracle vs Google: Can Oracle sue any company who is using Java and making profit?

In the javadoc api license there is the following clause: License for Evaluation Purposes. Oracle hereby grants you a fully-paid, non-exclusive, non-transferable, worldwide, limited license (without the right to sublicense), under Oracle's applicable intellectual property rights to vie...

tl;dr you cannot implement just parts of the java api

it must be complete implementation and it must pass a validation test

Thomas Owens

How does OpenJDK fit into this, though?

It has the same API.

ratchet freak

clause 2a: (a) fully implements the Specification including all its required interfaces and functionality;

download.oracle.com/otndocs/jcp/java_se-8-mrel-spec/…

and 2c: (c) passes the Technology Compatibility Kit (including satisfying the requirements of the applicable TCK Users Guide) for such Specification ("Compliant Implementation").

Thomas Owens

Has the text been the same? Android started with a much earlier version.

Java...5 or 6, I think.

ratchet freak

yeah from java 5: docs.oracle.com/javase/1.5.0/docs/relnotes/license.html

> i) fully implements the Spec(s) including all its required interfaces and functionality;

> (iii) passes the TCK (including satisfying the requirements of the applicable TCK Users Guide) for such Specification.

Thomas Owens

Interesting. But what about OpenJDK? Which is GPLv2 + Classpath Exception.

Also, all of that assumes that you can actually copyright an API.

I think that's what one of the questions is, isn't it?

ratchet freak

3:01 PM

clasue 4 in openjdk.java.net/legal/gplv2+ce.html

gpl doesn't really seem to be compatible with the javadoc license

which is openJDK's fault for picking the license

Thomas Owens

Doesn't the ability to copyright an interface destroy the ability of people to cleanroom reverse engineer proprietary software and build open implementations? For example, the open source software that can read Microsoft Office files, or MATLAB files. The interface is in a different form, but if I own and control "public String toString()", then that's kind of bad.

An interface is still an interface, regardless of if its a series of function declarations or if it's a structure of a file or data stream.

ratchet freak

frankly Google can argue that the java interface has been licensed under GPL+classpath because oracle owns openJDK

Ixrec

frankly, it's all a big lawyerfight we should stay out of

Thomas Owens

Again, it goes back to trying to explain to a bunch of people how software and software licenses work. It's hard enough to get professionals to understand that. :\

@Ixrec I don't really agree with that. We are where we are now with convoluted copyright and patent laws because the people making those laws didn't understand what they were regulated. All professionals should be inserting themselves into the process of regulating things that impact their profession.

Ixrec

oh, if there's a regulation thing coming up then yes

but that's a completely different debate from trying to figure out what you could get sued for today

Thomas Owens

3:16 PM

And this case has the potential to be devastating, depending on how they determine copyright applies to software.

gnat

3:39 PM

7

A: Where does my tool question go?

Question is About How to Use a Development Tool If the tool that you are using is a tool that is used to design and develop software and is used primarily in a software development environment, you should check out Stack Overflow. A number of development tools already have questions and answers...

> If your question involves how to integrate a tool or type of tool into your workflow or process, then it is likely to be on-topic on Programmers. Generally, if you aren't asking about how to use a specific feature of a tool, but trying to determine how to use the tool to achieve a well-defined goal, start here.

Thomas Owens

@gnat Yes. I was tossing the idea around in my head about questions like "how do I do X in Y", where "Y" is some professional-grade tool: DOORS, Rhapsody, Enterprise Architect. Things that programming enthusiasts would likely not own. But most of those should just go to the support forums for that tool.

Robert Harvey

@ThomasOwens I hesitate to remove Algorithms and Data Structures from the on-topic list, because I'd really like at least one bullet to say "yes, we really do answer nuts and bolts questions here, not just management, methodologies and SOLID."

Speaking of management, is SCRUM all that and a bag of chips, or is it Just Another Bondage and Discipline Management Technique?™

Ixrec

3:57 PM

depends on who's implementing it

I'm convinced a lot of the flak directed at various flavors of agile is about what happens when bad managers use it as an excuse to be bad managers, which they can do just as well with any other trend or methodology

Robert Harvey

So basically same as it ever was. New management technique flavor of the week abused by management for fun and profit.

Ixrec

and there's always people being overly dogmatic about it

starting with scrum and adjusting it to fit our reality has worked great for us, but it definitely needed adjustment

Robert Harvey

@Ixrec Like everything else in Software Engineering. Dogma seems to attract programmers.

People who desperately want the world to be black and white, and it's never gonna be.

Ixrec

the fact that a tremendous amount of our work is either blocked on other teams, consists of nagging other teams, or consists of researching what other teams can and can't do being one of the biggest things normal scrum doesn't really account for

for the moment we're basically putting non-user stories into our sprints, because otherwise more than half of what we spend development time on would be completely untracked, which would sorta miss the point of using a bug tracker and sprints and everything

Robert Harvey

Our company will be hiring a Project Manager. My first question to my boss was "Is he going to help me stay unblocked so I can get my job done more productively, or is he going to put in a bondage and discipline methodology that has so much administrative overhead that I can't get anything done?"

The truth is probably somewhere in the middle.

@Ixrec That's an eye-opener. You're tracking scrum points on administrative/communication overhead?

I could see internal user stories, like "we need this custom tool to get our work done more efficiently." But user stories on "soft" problems?

I figured you just padded your story points on your ordinary user stories for that sort of thing.

Ixrec

4:05 PM

it's not like the tickets say "talk to X about Y", it's more like "figure out if X is possible" or "deal with all the bugs we found in Q last week"

Robert Harvey

But the bugs are already in a tracking system, no?

Ixrec

or "migrate X over to the new shiny to unblock all those other user stories"

Robert Harvey

Research and Development is a legitimate story.

Ixrec

but bugs don't go into sprints, sprints are for user stories! supposedly

Robert Harvey

Pfleh. Bugs go into sprints.

Ixrec

4:06 PM

everything's legitimate if your goal is to get sh*t done

Robert Harvey

And everything legitimate is worth tracking.

Dodzi Dzakuma

I am trying to word a question for the site. My question is similar to "Is there a correlation to the number of tables (data storage) locations needed for a feature of a program to work to whether that feature is doing too much?"

The background is, I am trying to refactor a program, but there are places where I believe one feature of the program is trying to do too much.

One feature of the program has to reference 47 or so tables to work.

Ixrec

depends on your definition of "feature", "table" and "location"

but if you ask about this specific feature and the specific tables it's using, that might work

Dodzi Dzakuma

I feel like that is too much, and there needs to be some refactoring done (either to the tables information architecture or the program itself).

Ixrec

I have to know what kind of "feature" gets 47 tables to itself

Dodzi Dzakuma

4:15 PM

@Ixrec thanks for the advice. I'm having trouble finding any information on this on the internet. Even in books, I'm having trouble finding anything to go on.

My wording is also pretty bad because, I don't quite understand how to pose the question and the points that need to be considered.

Hold on while I figure out what I mean when I say "feature" and "table" or "location".

Robert Harvey

@DodziDzakuma The way you define "too much" is "is this program maintainable and readily understood by others?" If it isn't, you change the design so that it is. Also, the designer of your program has been watching too much Star Trek.

Dodzi Dzakuma

@RobertHarvey LOL. The person who made that video watched too much Star Trek!

That's a good one.

The program is NOT maintainable or understandable.

I have FINALLY convinced them to migrate from VB6.

I got them doing automated deployments and builds.

I also got them using version control.

Robert Harvey

That's a good start.

Dodzi Dzakuma

Now that we are transitioning to .NET, I need to quantify my findings about our difficulties in moving each screen to .NET.

Ixrec

that's a lot of good starts

Dodzi Dzakuma

4:22 PM

There are some screens that are referencing waay too many tables.

The names are cryptic, so I did a content audit of each screen.

I just need to be able to make a case that we should slowly start refactoring the database OR consider refactoring the database after we transition to .NET.

Robert Harvey

@DodziDzakuma You can't quantify this stuff. It's like asking "how many lines of code did you write today," and using that as a metric to determine programmer productivity.

What you can do is use sensible design practices.

Dodzi Dzakuma

One screen needing 47 database tables to function is smells of poor information architecture.

Ixrec

I wrote -60 lines today =)

Robert Harvey

@DodziDzakuma Of course it does. It also smells of a designer who doesn't understand database design.

Dodzi Dzakuma

@RobertHarvey I completely understand what you mean and I agree 100%. The CFO does not...

Ixrec

4:25 PM

the CFO is probably wondering whether this results in any improvements the end users would care about

Dodzi Dzakuma

Hmmm...

Robert Harvey

Are you the IT manager, or is someone else? It's his job to convince the CFO, not yours.

Ixrec

the usual argument is something ephemeral about "developer productivity" which of course is another unquantifiable thing

Dodzi Dzakuma

The CFO likes to dictate a lot of control over IT. My manager is not interested in change. I would be the senior developer of the project.

Robert Harvey

The nature of UI design and human behavior suggests that there isn't a UI form that demands more than about 8 tables, unless it's some sort of enormous dashboard.

Dodzi Dzakuma

4:26 PM

IT on this system consists of two people. Me and my manager.

Robert Harvey

Is your resume polished?

Dodzi Dzakuma

@RobertHarvey Good stuff!

Ixrec

to be honest, if I introduced version control, automated builds and VB.NET to my company, I'd be tempted to quit while I'm ahead and just write code as far away from the legacy bits as possible

Dodzi Dzakuma

I'm on linkedin.

@Ixrec I kind of agree with you, but I think there is still a lot of room for me to grow as an individual in the IT field here.

Robert Harvey

It's not necessarily a bad thing to work for a "garage" company. I did so for many years. The owner had me write an entire practice management suite in VBA and Access. You can make it work, if you're careful. But after I left, he hired a bunch of people and converted it to ASP.NET.

Dodzi Dzakuma

4:28 PM

Because it's just me an my manager I get to wear more hats.

That equals more experience.

Robert Harvey

That's fun, isn't it?

Until you have to do customer support also.

Dodzi Dzakuma

Do the WHOLE thing...

@RobertHarvey going back to your comment on the nature of UI design... are there any studies on that?

I'd love to read up.

Robert Harvey

Experience, not studies. Bits of "best practice" that have been gathered up over the years. The state of the art is probably MVVM.

Usually, the inexperienced person uses one table and one form with 300 controls on it. What happened here?

Ixrec

speaking of customers, did I ever tell you guys about the one our helpdesk said "does not care ... if the data is inaccurate, he does not want to [upgrade] as it will change the font"?

we got a good laugh out of that one

Robert Harvey

@Ixrec WHO MOVED MY CHEESE???

Dodzi Dzakuma

4:31 PM

@Ixrec You made me spit out my coffee.

Thomas Owens

@RobertHarvey I'm torn. On one hand, any good algorithm question for us would essential be a design question (as far as I've reasoned through it - if I'm missing an example, please point it out). On the other hand, being explicit is helpful so people don't get discouraged.

Dodzi Dzakuma

@RobertHarvey Who moved my cheese is a great book!

@RobertHarvey "One form... 300 controls!" Plus a circular dependency in that control information is read from the database, so if it isn't in the database first, the program dies before it can produce any useful information.

But you've nailed it.

Robert Harvey

@ThomasOwens Ana is wrong about the four bullets. That's an arbitrary, unsubstantiated number, just as arbitrary as 3 being the maximum number of custom close reasons (when Stack Overflow has 5).

@DodziDzakuma That's a designer with no design experience. The only way you combat that is to show examples of other programs that have sensible design, and demonstrate how well they work. Even then, there might not be any interest. At the "garage" I told you about, most of the development innovation was mine; I just did it.

Dodzi Dzakuma

@RobertHarvey I actually that kind of power here as well. Creation of a logging program. That happened. We can now trace problems as well as get early notifications on major events.

Robert Harvey

@ThomasOwens Ana wants a specific site scope. The way you do that is by being specific. We can do that without the number of bullets being onerous, but it is not four.

Dodzi Dzakuma

4:35 PM

My manager asked me to turn it off for him because he was getting too many messages.

Robert Harvey

@DodziDzakuma Then don't send those messages to your manager. You've already established that he's Shlemiel the Painter.

He doesn't give a shit.

Shlemiel gets a job as a street painter, painting the dotted lines down the middle of the road. On the first day he takes a can of paint out to the road and finishes 300 yards of the road. "That's pretty good!" says his boss, "you're a fast worker!" and pays him a kopeck.

The next day Shlemiel only gets 150 yards done. "Well, that's not nearly as good as yesterday, but you're still a fast worker. 150 yards is respectable," and pays him a kopeck.

The next day Shlemiel paints 30 yards of the road. "Only 30!" shouts his boss. "That's unacceptable! On the first day you did ten times that much…

Dodzi Dzakuma

@RobertHarvey @Ixrec Thanks a lot for your help and the information you gave me. I'll use that as a basis to start searching for more material to make a case.

I also realized that my question would have been a bad one.

No real good answers.

Robert Harvey

"...and he was enlightened."

Thomas Owens

@RobertHarvey I agree. I think that 7 is a reasonable number, though.

3

Dodzi Dzakuma

@ThomasOwens @RobertHarvey The number 7 actually matches the data I compiled from my audit of the program.

There are a large number of forms that only reference 7 tables.

There are other forms that reference one too many.

Might be on the way to some validated research here: books.google.com/…

Robert Harvey

4:50 PM

7 is a magic number. Combine it with Ana's number and you have 47. Oh, wait.

@ThomasOwens Algorithms and Data Structures is an important category. It states, by example, that we care about more than just Architecture Austronauty.

Dodzi Dzakuma

I think that even an enormous dashboard can be simplified through the appropriate use of a database view.

Robert Harvey

@DodziDzakuma If the dashboard is composed of a number of applets or subforms that each use less than 7 tables, it still counts.

Dodzi Dzakuma

Yup!

Robert Harvey

But nowadays I think the fashion is to feed those things with something like a microservice.

If you're so inclined.

@ThomasOwens SE thinks that we can do it with four bullets, because that's how they did it on Stack Overflow. But Stack Overflow's scope doesn't arise from those four bullets; it is what it is today because of years of polishing and refining the scope on meta. Well, that and "What have you tried?" And 9 out of 10 questions being "fix my broken code."

So those four bullets do not in any adequate way describe SO's actual scope.

Thomas Owens

@RobertHarvey I think that may be part of the problem. :\ SO is so narrow. Yet they are still allowing things that should be on Programmers. Probably because they don't understand Programmers. Just the other day, there were some modeling questions on SO.

Robert Harvey

4:59 PM

@ThomasOwens You're still making the tacit assumption that site scopes shouldn't overlap.

Look at this question:

17

Q: Why was the register keyword created?

While reading Keywords That Aren't (or, Comments by Another Name) by Herb Sutter I came across these lines: That's right, some keywords are semantically equivalent to whitespace, a glorified comment. And We've seen why the C++ language treats keywords as reserved words, and we've seen t...

c++ c c++11 language-lawyer

Would people accept this question on Programmers? Or would they consider it too opinionated?

Right now we are ruthless with new questions. But that's because nobody understands the site scope, and closures are the only way we currently have to steer that scope. If we became a site that's focused on software engineering principles, I think more people would come here with their software engineering questions.

Thomas Owens

@RobertHarvey Some overlap is OK.

But I honestly hate opening up 5 tabs to look for questions about software development that I can answer.

The type of experts that should ultimately be attracted to a site won't do that.

Robert Harvey

How would changing Programmers to Software Engineering cause that?

Thomas Owens

It wouldn't. That's current state.

Robert Harvey

What 5 tabs? The ones for Open Source.SE, Project Management.SE, and (what's the other one?)

Thomas Owens

Stack Overflow, Programmers, PM, SQA & Testing, Engineering.

I also open up Workplace and Open Source.

For example, why is this question on topic on Stack Overflow?

Robert Harvey

5:11 PM

From my jaded perspective: it's a relief from the relentless onslaught of "fix my broken code" questions, and Programmers isn't currently a suitable venue for it.

Stack Overflow was never meant to be a "fix my broken code" site, but that's what it has become.

Thomas Owens

@RobertHarvey I like that. I have my own code to debug.

I don't want to fix yours, too.

But, on the other hand, there are still interesting questions on SO that I'd be interested in answering.

Robert Harvey

The problem is, you can't find them in the vast sea of "don't work, how fix."

Thomas Owens

Well, it's a little easier with New Nav. Unfortunately, I only have new nav set up at work. I have a tab for new questions with tags like "requirements" nad "uml".

Robert Harvey

I still hold out hope that Software Engineering can eventually attract those people, the ones who would like to ask more "conceptual" software development questions, but who don't want to toss them into the SO sea.

It doesn't do that now, because we're too busy closing "halp" questions.

Thomas Owens

Closing "halp" questions where? SO closes their fair share of bad "halp" questions.

Aaron Hall

5:16 PM

Wait, .... wait did they decide to change the name?

Robert Harvey

@ThomasOwens Well, I mean any question that asks "This code doesn't work, how do I fix it?"

Which is pretty much 90 percent of all questions that are asked on Stack Overflow.

Thomas Owens

I suppose I could start flagging all of these good Programmers questions for migration. But then the SO mods would hate me. Or just appoint me SO mod to migrate all the good ones that I can.

Robert Harvey

38

Q: New Site Name and Scope Proposals

From Ana, a Stack Exchange Community Manager in the previous question about changing the site name: It took us a good while, but the Community Team has circled up and here's where we stand on your request to change Programmers.SE's name. We agree that renaming this site is a good idea....

discussion featured on-topic-definition scope site-definition

@ThomasOwens They're not off-topic there.

You're still thinking in terms of "neat, tidy categories." SE has never worked that way.

@ThomasOwens Stack Overflow's users don't recommend good design questions to Programmers. They keep those to themselves. They recommend those questions to Programmers that they don't want, for the same reasons that we don't want them either.

Thomas Owens

@RobertHarvey They aren't off-topic, but I'm saying that maybe they should be.

Robert Harvey

You could float that idea on Meta.SO, I suppose. I doubt it will get any traction, though.

Thomas Owens

5:20 PM

It's just annoying to find good requirements and design questions on both SO and Programmers, where you could copy and paste the same question on both sites and get good answers.

On one hand, increasing the scope of SO is bad because it will just make it harder to find the questions that you can answer. New Nav helps, but I don't think SO has scaled well.

On the other hand, I'm missing many good questions that I can answer because they get lost in that noise.

Robert Harvey

The problem, in a nutshell.

Thomas Owens

Let's take some kind of architecture or design expert. Let's say...Martin Fowler. He wrote a great UML book. He may watch Programmers if he was interested in finding good UML and design questions to answer. He probably won't watch SO because those questions would be buried in fix-my-code questions.

gnat

Yeah, getting rid of the overlap is probably the wrong direction to be going. It'd be like trying to de-overlap ProgSE and SO. The topics overlap; accept it and mark the natural boundaries. — Shog9 ♦ Oct 28 '14 at 20:59

Thomas Owens

I thought the idea behind SE was to attract experts to answer questions. Experts don't want to wade through overly broad scoped sites to find the handful of questions they want to answer.

Aaron Hall

I don't know if that Python question is good, but my answer just got an accept.

Thomas Owens

5:26 PM

Natural overlap shouldn't be avoided. Artificial overlap should be.

Robert Harvey

Can you ask the question in a way that doesn't provoke an endless parade of opinions, requires a book chapter to answer, or asks us to make a big list of things? — Robert Harvey 1 min ago

@ThomasOwens Stack Overflow is too broadly-scoped, but that doesn't mean that Software Engineering has to be. We can refine our scope without Stack Overflow's help.

2

Thomas Owens

5:47 PM

@RobertHarvey That's true. But still doesn't address my concern that the experts on things like architecture, design, and methods are going to come to Programmers because it's a refuge from "fix my codez" questions. Many people who need help are going to find SO (by knowledge, by Google) and see not only no indications that their questions are off-topic, but that similar questions get answers. But when they start asking these questions, the most enthusiastic experts won't be there to give help.

Robert Harvey

Oh, I don't know about that. Eric Lippert enthusiastically answers questions here on Software Engineering, when people ask good ones.

Thomas Owens

That is true. But I think he's the exception, rather than the rule. He does participate on both sites. However, I know many people (who aren't at Eric Lippert's level of expert in their field) who won't bother wading through Stack Overflow, though, to find nuggets. They'll come here and that's it.

Robert Harvey

Exactly my point.

Thomas Owens

Isn't that doing a disservice to askers, though? By permitting their architecture, design, and process questions on Stack Overflow?

Robert Harvey

We don't have any control over that.

I agree with you, but there's nothing we can do about it.

Thomas Owens

5:52 PM

I don't know about that. As a community here, true. There are other avenues, though.

Most of us have SO accounts, I think. Plus, I hope that SE would realize that this is a problem that is doing a disservice to people seeking answers to their questions and iron-fist the hell out of a solution.

Robert Harvey

The phrase "when pigs fly" comes to mind. :)

Thomas Owens

Probably. I do think that Step 1 is a name change.

And, hopefully, a theme change.

Robert Harvey

The name change alone would make a huge difference. I used to didn't think so, but the change to Electronics is like night and day.

Granted, that may be correlation without causation, but it seems remarkably coincident.

Thomas Owens

Want to know the other benefit? Easier to explain the site to people. People know what "Software Engineering" or "Software Development" is and can think of questions to ask that would probably be on-topic, even if they were too broad or opinionated.

But people don't know WTF to ask on a site called "Programmers". Is it about what to name a cat?

Robert Harvey

All true. It's gotta be "Software Engineering." Whether that term really exists or not, it hints at dress shoes, not flip flops.

People have to come in here with the expectation that we will demand a certain level of professionalism from their questions.

Transcript for