Anselm Lingnau wrote: > Alessandro Selli wrote: > >> When I train people to LPI certification objectives I train them to >> pass LPIC-1 certification exams, not to be good, all-round GNU/Linux >> sysadmins. > That's the difference between the two of us then.
Not really. This is the difference between what it's good to do and what it's possible to do in the given circumstances. > At Linup Front GmbH, we > don't »teach to the exam«. The goal of our training materials (and by > extension the classes we teach based on them) is to give people what they need > to know in Real Life. This what I can do sometimes when I teach non-LPI related, on-line classes, i.e. courses that are tailored to the specific needs or desires of the attendant with no certification involved. Been this kind of people generally new to Linux, then I have a lot of room to propose changes to the schedule, both in terms of covered objectives and time. Alas, this is not the case when I am asked to teach a guy "what he/she needs to get the LPIC-1 cert. in 40h, including an exam test session". There are cases when I should refuse to do it because I understand the person is just not ready and will not be in a month, but I just cannot. > In many cases this quite deliberately goes beyond the > LPI exam objectives. Again, this is your decision. I am happy to learn you can afford it. I envy you. Really. So far this year I could not. I have a base of 28 sets of slides that form the base of the documentation I use for the various Linux-related courses that I am asked to deliver. Some have sections titled "Good practices and gotchas", or "Security implications". I cannot remember using them once in LPI-related courses this year, because doing so would entail skipping or at least reducing to below the bare essential something that they might be asked when taking the exam. When I have to choose (and I must do so every 15 minutes or so during a course) I must give priority to the exam stuff, because that's what I am asked to do and - they never forget to remind me - what I am paid to do. > For example, in our introductory curriculum we cover shell programming in > great detail, illustrating various programming and debugging techniques, even > though it is possible to pass the LPI-102 exam with a fairly cursory knowledge > of the syntax of various shell constructs. A few times I suggested people with a clear interest in scripting to take a specific course or a non LPI course, because covering scripting in this detail would itself take about half the time we had available. > This is because I don't see much > point in drilling people on syntax rules I don't! Most of the times I run examples and provide with a table of "Most important switches and parameters" or "Fundamental Syntax Rules". However, most LPI courses I hold are not tutorials, the are not designed to teach new concepts and technologies one needs to know to land a job or to pass a professional certification. Would one need such a course, because he/she starts from a LE level or worse, I estimate the required time in a month, if the person is capable and motivated. I remember the first professional course I took, Data Transmission Technologies, was a 500 hours course and lasted over three months. That's what you need to learn new stuff. Instead, LPI courses I hold end up expounding the covered exam objectives with examples of what is expected of a professional entitled with the given LPI certification. That is, given that I cannot teach in a week or less what one needs to know to pass an exam, I provide the guy with a measure of his/her already attained proficiency in the covered subject matters, so that he/she can estimate the chances he/she might pass the exam the following week, or the job needed to catch up. This catching-up requires a learning-oriented course, that we offer as a separate package of Linux courses what do not have LPI in the name, because they are not tied or limited to LPIC exam objectives. Needless to say, the ratio of people taking these courses to those who wish to get an LPI certification after having successfully installed Ubuntu in a virtual machine under Windows is extremely small. For this reason, the project of offering long-term LPI courses has not seen the light. > if they don't know what to do with > them, and so I'd like to leave them with enough knowledge to write and debug > simple shell scripts that actually do useful things, and to puzzle out more > complex ones they may encounter. Do you really manage to do this in a week when you have to cover LPIC-1 related objectives? > That sort of thing does take time but it is > usually time well spent. It's time I just am not allotted. This is the problem. It's not the LPI objectives, it's not GNU/Linux (well,... :-), it's not the usefulness, real or perceived, of ed as an regexp testbed before sed unveils the splendour of the streaming parsing world. It's the: "I am to do it in a week. Beyond that, I should take a leave from my employer, because they won't pay for the extra time needed". > This approach also acknowledges the fact that, at the > end of the day, many of our customers are interested in learning Linux rather > than passing exams; That they are just interested in passing an exam or not, I *always* tell them, on the first day: "You are really only going to learn Linux if you use it, and your proficiency in using it will run in the same measure you will use it to do everything you need to do with a PC (plus an extra couple of things you never though doing with your computer before)". I once was asked by a guy I was explaining ARP and routing to, if I could also explain how Tor works and what techniques are most used to snoop TCP/IP traffic (I noticed he had Kali Linux installed in his laptop, but he could evidently do little of it). I would have jumped of joy to be able to. But I could not. It's one of the things we could only do in our spare time, but he came from out of town. So the answer was just a "No, sorry. It's not in the objectives and we'd run out of time for the wireless stuff." But I know it's not only a very interesting, exciting and fun topic, but one that could also land a good job in the IT security field. I could set up a separate course on the matter, but I know my employer would not be any interested because we were never asked to deliver such a specialistic course (=what money could we make from it?). [...] > As far as time constraints for classes are concerned, boo hoo hoo. We also get > asked to teach LPI prep classes in completely absurd time spans (e.g., 2 days > for LPIC-2). If that happens we make it clear that there is no way to fit 5 > liters of water into a 1-liter container, no matter how fervently the customer > wishes that to be possible. Everything else would be dishonest, unfair to both > the instructor and the class participants, and, in the long run, bad for > business. Of course this is true. Of course I am asked my opinion on the requirements of customers. But my opinion is not binding, it's only consultative, as the decisions are taken by Management. And of course to them the balance sheets come before a good, sound, inclusive and all-round Linux tutorial. I cannot blame them, I know we had to go from 50 to 30 full time employees in the past four years, I know they'd be happy to sell such courses by the dozen every week. It's just that we, as well as our customers, cannot afford to. > This approach usually leads to more realistic class-time > arrangements. So far we've had a take-or-leave choice. And so we taken them, warning of the related risks. So far we were lucky, no one has smeared us on the social sites for not delivering the "promised" LPIC certifications or for been incompetent or, worse, fraudulent. We wish we could do better, I wished I could have a room with equipment devoted to experimenting and hacking, not just delivering courses. Because it's doing that that teaches you things, and it's what many people would just love doing but cannot do at home or at work or even in their University. But we too cannot, the generated return would not justify the required investment. Classrooms generate more return when they are devoted to ECDL courses and exams. [...] > If you're using our books I am not. So far every LPI course we've hold was in Italian, and I produced our own material for English, non-LPI Linux courses we held. [...] > Is Classless IN-ADDR.ARPA delegation part of the LPIC-1 objectives? > Not as far as I'm aware of – and since LPIC-1 doesn't deal with server-side > DNS at all, there is no reason it should be. I mentioned it as a counterweight > to the claim that network classes were important for reverse DNS, which they > aren't; they just make certain things more convenient. I did not advance that claim. It was Bryan J Smith who wrote «Because of how IANA assigned IPv4 blocks and how the authorities are used, this is _why_ in-arpa.add r zones (reverse DNS records) are required and how they work, on the IPv4 classes. ;)». [...] > In fact we offer a more in-depth DNS class (including details of > topics like DNSSEC which are not part of the LPIC-2 exam beyond > »knowing what it does«) to people who are interested in actually > running a real-life DNS server as opposed to just passing LPIC-2; Again, both I and many of the people who attend our LPI classes are interested in teaching/learning more that the strict necessary to pass exams. It's just that we are not allotted the required time to do it, and are both given strict time requirements by people either unable/unwilling to pay for the right learning process, or who dramatically underestimate the needed time to advance a good technological knowledge in the matter, or both. I was called an idealistic guy so many times, I'd be rich if I got an Euro cent each time. Regards, Alessandro _______________________________________________ lpi-examdev mailing list [email protected] http://list.lpi.org/cgi-bin/mailman/listinfo/lpi-examdev
