Just how do our peers see us?

Posted 28 Jun 2002 at 19:14 UTC by Chicago Share This

alternative title: Is my qualification worth anything?

Following up from a suggestion by fxn about who to certify and why, and a thank you to mattr who keeps in his notes his reasons for certifying.

The question he asked was why are we certifying people? And I feel that that is almost more important then actually seeing a certification. Why am I certified as a journeyer? (and I dont mean the actual reason, I mean what made them tick the box journeyer).

Are we certifying people becasue they are good at programming? Or are we certifying people because they are helping the community? Or becuase they have produced the most programs? Or the length of time they've been around for? Currently I suppose I loose on all these aspects, but I'm learning.

Personally, I see myself as an apprentice. Of course this is by the standards of the people who use this. Compared to the average person on the street, they are unable to compare the difference between some of us. What I am interested in is bytesplit certified me as a Journeyer. Not that Im complaining to have lost my green colours or anything, but still...

However, I must object strongly to the accusation of doing computer science. I am not, nor ever will be, a computer scientist. Why? Well look at the people who are doing the computer science course? Half of them dont have a clue. Certainly, they have no ... feel for the computers. They dont seem to care about them. And because of that, the general "Computer Science" course is becomming slowly usless. No, I am an Engineer, and even though we share certain modules with the computer science students, I feel that this course is a much better use of university time. Im sorry, but I really dont think that anyone with a Computer Science degree has actually done anything worthwhile. Now perhaps Im wrong about that, but after just finishing my first year at UEA, and not having covered anything that I hadnt covered at Alevel and in my own interests, I feel that I have aceademically wasted a year of my life.

And its interesting reading the pages of other peoples. Those out there who are certified as "masters" who have PHd's, and surley, surley they must have some affect on the universities who are causing the courses to be set. Because at the end of the day, I take into a bigger consideration what people in Advogato say about me then I do about a piece of paper saying that I've got a degree.

Especially as once I have these certifications from you, and if the system was changed to force people to give reasons why they had certified, then perhaps I would learn more, faster. And perhaps we can solve some of these computing problems better.

So are qualifications in computing and IT worth anything compared to the certification of our peers? And if so, is there a relationship between the official cirtification and the levels on advogato?


Reply, posted 28 Jun 2002 at 20:26 UTC by salmoni » (Master)

I can see your point - I am certified as a journeyer too, but see myself firmly as an apprentice, despite many years of experience in programming. I would rate myself as an apprentice because although I've been programming for almost 20 years, it is all self taught and piecemeal stuff. I have a project on the go which is going well (I think!), but I imagine that anyone with half a brain and a lot of motivation could do as well if not better.

Maybe motivation is the point though - we at Advogato are motivated to achieve things that we don't necessarily get tangible rewards for - sure we get status, an education and community acceptance, and maybe even praise from people outside (which is fantastic btw), but I have seen no money for all the work I have done. If I had devoted the hours to working in McDonalds, I would be a lot richer.

Perhaps the motivation we show through our journal entries and the work we produce is the evidence that people use to judge us? After all, an active contributing member of the community is worth more than ten people just passing through?

I'm rambling a bit, but that still leaves the problem of whether the certification is worth anything? I think so - we have chosen this community to be part of our lives and can leave it at any time we want. The only requirement to participate is to participate, and appreciation of our work is a great spur to a) produce more, and b) ensure that the work we produce is of a higher quality than if we were cubicle jockeys cranking out code for a fat monthly paycheque, so maybe a few seconds of someone elses time is worth the effort, if not for us, then for the wider world.

>So are qualifications in computing and IT worth anything compared to the certification of our peers?

Yes, because people outside of Advogato can relate to them ;) I might even put my certification on my CV!

Reply, posted 28 Jun 2002 at 20:26 UTC by salmoni » (Master)

I can see your point - I am certified as a journeyer too, but see myself firmly as an apprentice, despite many years of experience in programming. I would rate myself as an apprentice because although I've been programming for almost 20 years, it is all self taught and piecemeal stuff. I have a project on the go which is going well (I think!), but I imagine that anyone with half a brain and a lot of motivation could do as well if not better.

Maybe motivation is the point though - we at Advogato are motivated to achieve things that we don't necessarily get tangible rewards for - sure we get status, an education and community acceptance, and maybe even praise from people outside (which is fantastic btw), but I have seen no money for all the work I have done. If I had devoted the hours to working in McDonalds, I would be a lot richer.

Perhaps the motivation we show through our journal entries and the work we produce is the evidence that people use to judge us? After all, an active contributing member of the community is worth more than ten people just passing through?

I'm rambling a bit, but that still leaves the problem of whether the certification is worth anything? I think so - we have chosen this community to be part of our lives and can leave it at any time we want. The only requirement to participate is to participate, and appreciation of our work is a great spur to a) produce more, and b) ensure that the work we produce is of a higher quality than if we were cubicle jockeys cranking out code for a fat monthly paycheque, so maybe a few seconds of someone elses time is worth the effort, if not for us, then for the wider world.

>So are qualifications in computing and IT worth anything compared to the certification of our peers?

Yes, because people outside of Advogato can relate to them ;) I might even put my certification on my CV!

problems with the trust metric, posted 28 Jun 2002 at 23:56 UTC by jbuck » (Master)

There are huge problems with the Advogato trust metric, because you are asking people to make a judgement on a fuzzy metric, using definitions that in some cases contradict the words. For example, by a strict reading of the rules, we couldn't even rate Dan Bernstein as an Apprentice, because despite his enormous skill as a software designer, his software is not free software. Thus we see divigergent ratings for some folks based on whether people use the dictionary meaning of the word, their intuitive sense of what the word means, or if they read the Certification page carefully, or if they just rate all their friends highly.

That is, Raph picked words that are associated with describing skill generally, but picked definitions for those words that limit their application to free software. Someone who writes proprietary software but who occasionally sends in bug reports to free software projects rates as an Apprentice, at most.

Finally, there is the implicit assumption that trust in a person's judgment and maturity increases along with that person's programming skill. There are a number of cases where people on Advogato who are clearly Master-level based on either sense of the word (mastery in terms of skill as well as contribution to vitally important free software projects) are rated as Apprentice by someone else of similar skill, just because the two people don't like each other.

Raph's definition of "Master" requires the person to work full-time or equivalent on free software. Many advogato members ignore that definition and rate people as Master based on their judgment that someone is a really good coder or some such. I rated myself Journeyer because of the way Raph defined the categories, but others, using the dictionary sense of the word, rate me a Master. So it goes.

Certification, posted 29 Jun 2002 at 16:54 UTC by jul » (Master)

I think trust metrics is an interesting system, because people often judge themselves more severly than their peers do. As an exemple, auto-notation of students is -in average- a fifth lower than what teachers do. I guess the trust metrics measure what people would like to see rewarded. We are a community, and the grade might measure the level of value someone thinks you bring to everyone (him especially). What is interesting is not what trust-metrics is supposed to be, but how people use it anyway.

So trust the people who think you are a <whatever>, and give them back their trust in trusting yourself.

Think positive.

Certification, posted 2 Jul 2002 at 09:56 UTC by pom » (Master)

The princip of the certification might be good, but its actual version is quite poor, because of the fuzziness of the notion it covers. Even if there are guidelines, they poorly fit to the common intuitive meaning given to levels.

First, there should be a distinction between people certification and project certification: someone may work on projects of very different quality and difficulty, many programmers may work on a same project without having similar skills.

One of the most important drawback in the free software community is the lack of a refering process for the projects, allowing a first classification on the multitude of the projects (difficulty, quality of the code, openess, usability, aso.). As a result, a 100 lines source project performing an easy common task may not be distinguished, at first glance, from a 100000 lines source one implementing a very innovative approach for a very difficult problem.

Refering is a time and energy consumming process, which falls in the same kind of "boring" work as documentation and administration tasks: it has to be done.

Last, the certification level of a person could be related to its refering activity as well, in terms of comments quality, objectiveness of the appreciation and difficulty of the done refering.

You are mixing up Computer Science, Engineering, posted 2 Jul 2002 at 19:06 UTC by mdanish » (Journeyer)

However, I must object strongly to the accusation of doing computer science. I am not, nor ever will be, a computer scientist. Why? Well look at the people who are doing the computer science course? Half of them dont have a clue. Certainly, they have no ... feel for the computers. They dont seem to care about them. And because of that, the general "Computer Science" course is becomming slowly usless. No, I am an Engineer, and even though we share certain modules with the computer science students, I feel that this course is a much better use of university time. Im sorry, but I really dont think that anyone with a Computer Science degree has actually done anything worthwhile. Now perhaps Im wrong about that, but after just finishing my first year at UEA, and not having covered anything that I hadnt covered at Alevel and in my own interests, I feel that I have aceademically wasted a year of my life.

You are mixing up Computer Science, Engineering, posted 2 Jul 2002 at 19:29 UTC by mdanish » (Journeyer)

I accidently clicked the mouse button on this laptop and that posted only the blockquote of a paragraph from the article, somehow. Anyway, in brief, my response:

Computer science is not engineering is not systems management. Computer science is not about dealing with the peculiarities of the system-of-the-month. It's a broad topic dealing with logic, mathematics, processes and state. I would hardly categorize computer science by the people who graduate from it, most of whom are looking for the quick buck and easy jobs they heard they might get. Those types aren't interested in actual computer science.

If you felt the course was too easy, then you should've taken a more advanced course (and if you couldn't, either have tried to test out, or suck it up? I've had to take stupid courses too, to get to the ones I wanted). If you felt that the course taught you nothing useful, perhaps you are in the wrong field. While it would probably be useful to have an exposure to computer science topics, no one is going to force you to if you really just want to learn how to manage Unix systems or put parts together or something.

Computer Science, posted 2 Jul 2002 at 22:18 UTC by pom » (Master)

Computer Science is quite a fuzzy term; In its broad meaning, it covers all sciences dealing with information processing. Depending on the actual definition used, some works may fall either in computer science, mathematics, or engineering category.

For instance, signal processing and numerical filtering is often considered in both computer science and engineering, automatic schematic drawing from abstract datas may be considered both as computer science and mathematics.

Computer science courses also handles many different aspects, depending on the choices of the scientific board; It may go from microelectronic (hardware implementation of algorithms) to language theory (up to model theory, and various fields of formal logic), but also data structures for specific uses (hashes, graphs, structures for computational geometry, meshes,...) or even algorithmic arising from theoretical models of practical problems.

Software programming actually correspond to a "final" part of computer science, the same way that experiment is only part of the work of a physician. Some computer scientists never program, the same way some physicians only do theoretical physics; some others program a lot, and uses computers the way a writer uses its pen: to check ideas, to experiment new structures, aso.

The first computer science courses are, for programmers, as boring as the first course of a foreign language, for someone that knows its rudiments. But even, if you practice a foreign language well, litterature and poetry courses, which come later, may be of interest and so may be linguistic approaches. The same applies for computer science: even if you practice some programming languages well, you might find some interest in the study of data structures and algorithms.

Computer Science or Engeneering is not the point., posted 4 Jul 2002 at 08:35 UTC by jul » (Master)

"On internet no one knows you are a dog". The result is contributions are judged on their quality, their result rather than on the background of the contributor. The ability to take a stake in project relies rather on good sense and communication. Free Software, is about sharing the knowledge, and some projects are down by people without a clue in computer science and they do good job like SPIP. For those who can read french.
If we rely on Computer Science degree for certification we strongly get in opposition with free software community practices. For instance James Simmons who used to lead linux-framebuffer (maybe he still does) has a degree in Physics, therefore he would not be certified though he was (is) a project leader!

Subjective and fuzzy is fine, it is a web of trust, posted 5 Jul 2002 at 09:18 UTC by abraham » (Master)

I see the certification as "how much can I vouch for that persons contributions to free software". This means that someone who have written, or have made significant contributons to, a free application that is important to me and where I follow the dvelopment, will be ranked master. If the application or contributions are less significant, he will be ranked journeyer. If it is just a simple patch or two, he will be ranked apprentice. Documentation and intelligent advocacy also count as contributions in my view.

A problem is people on master level, whom I know for a few patches. Should I certify these as apprentices, because that is what I can vouch for? I'd prefer to do so, but now I typically leave them alone, I fear it might degrade their classification, or be insulting.

The official guidelines I only take as "hints" to establish the levels, there is no way I can know how much time people on general use on developing free software, I only see the product. Also, the rules don't take past contributions into account, if rms retired today I'd still want to list him as a master. And they don't take the significance of the contributions into account. By the rules, I'm a master because I work full time developing a free software program. However, it does not have a single user within the community. So it should not count. My present contributions are on apprentice level, but I believe my past contributions should put me on a journeyer level.

> Are we certifying people becasue they are good at programming?

I am not, not directly in any case. However poor programmers will have a hard time makign significant contributions.

> are we certifying people because they are helping the community?

I am, at least for helping my part of the community.

> Or becuase they have produced the most programs?

Only if these programs are significant to me.

> Or the length of time they've been around for?

Not directly, but I'm much more likely to notice someone who have been around for some time. Trust increses with time.

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!

X
Share this page