What mass/volume of this plant technology is required to recycle the CO2 for a single person? Also, does this technology have any consumables of its own such as H20? Seriously, do you think an office plant provides you with enough air to sustain your life?
As far as producing 02 and Water on Mars, how much equipment is required for that (and include the maintenence equipment)? Columbus was taking a risk that he would get to Asia before food ran out. He knew that he could buy his replentishment supplies in Asia. Furthermore, he knew that his trip back would take about the same time as his trip there (Asia/Europe didn't move closer/further as his trip progressed). He didn't plan on hitting a continent before Asia. What if he had hit a giant desert continent with no discernable food source? Ok, teach a man to fish... but the water problem remains unless they had decent destillers or something for getting drinking water from salt water. History is not my best subject.
By the way, it is going to take a lot of propellant to come back (Mars gravity is close to Earth gravity). That implies that you have to store it in those same discarded (oops) boosters that you took off of earth in (or replacements sent to mars on a seperate trip). We can't take off of earth effectively without a costly launch platform, what makes people assume we can do it on mars (because mars is all rocky, just like the moon so we can use the same launch technology)? How are we going to get that set up? Those air force pilots better bring along a couple of good constuction workers. Hell, just landing a craft upright will take a lot of fuel (or you need to build a big ass crane to upright it after it lands, which exacerbates the mass problem).
because she wants to know what her son is doing 18 hours a day tied to the computer
And you can't think of anything better to replace "I was downloading porn" than "I was programming?"
Sheesh.
I can't help bust ask "why"?[sic]
Answering that question would provide insight as to what to teach her. If she just wants to understand the draw of programming, there are probably better ways to make her understand than to teach her programming:).
In what context is Visual Basic too difficult? If presented as a blank slate followed by "and here's how to write a compiler in VB" I can see the difficulty. The materials used for teaching matter a lot more than what is being taught. Even HTML is a pain in the ass for beginners without a proper tutorial: "which way does the slash go?", "is h2 bigger than h1?", "are you sure I can't break the computer?"
I remember a SAM's publishing book that was a carryover from Peter Norton's Guide to Visual Basic (They changed the name after VB2 to Visual Basic 3/4/5/6 in 21 Days/24 Hours, something like that). I have used various versions of that book to teach "yes you have to plug it in first" types the basics of programming. One of the first exercises is a simple alarm clock (using a timer control and the system beep). Then a simple text editor with copy and paste. After that, you should provide your own fractal drawing sample (a simple recursive function with lineto's).
Using the above reference, VB is easier to learn than just about anything else (including HTML). It is/was easily the best book published by Sam's, and probably one of the best programming tutorials as well (and I have the little Lisper, My First C/C++ program, and a collection of other tutorials that I haven't had as much luck with).
Incidently, even if you are using a self running tutorial, make sure you are sitting nearby and well versed in first aid. I've seen people come close to panic attacks when the IDE crashed on them.
Teachers fear technology replacing them, but they only have themselves to blame as they did not keep pace with it.
I had some good teachers (and some poor teachers) when I went through school. My parents are firm believers in the public school system, but they always took care to undo some of the misinformation that is taught in public schools. My faith in the public school system isn't nearly as strong as my parents'.
I spoke to a girl in a teaching credential program a few months ago. She said something to the effect of, "I would be a great teacher. I'm terrible at math. I'd tell them that they don't have to learn it because, 'hey, I turned out fine!'"
My mother is on the school board in my home town. She had to explain to teachers the concept of negative numbers (they were faced with budget cuts because they had a deficit). She had to make an analogy to credit cards before the teachers understood. Grant it, these were not math teachers and were at the k-3 school, but seriously, WTF?
Another story, the "math guru" teacher at the same school was at a school board meeting complaining about the uncompensated extra hours teachers put in. His example: "I have a high school algebra textbook that I keep by my bed that I sometimes read FOR FUN." Again, WTF? A teacher who dubs himself a "math guru" doesn't know algebra? I actually had this teacher when I was in 3rd grade for math (18 years ago). I did learn my multiplication tables from this guy.
I was a big geek in elementary school (ok, still am). I remember arguing with a science teacher because I knew that there were more states to matter than solid, liquid, and gas (such as plasma, superfluid, superconductor--I only knew of plasmas at the time). I actually had to argue that since a gas undergoes a second order phase transition (didn't know what that was, but I had been told about it), that it was a different state of matter when it became a plasma. I ended up getting sent outside for disrupting the class. All this because I asked my dad (a plasma physicist) what he did for a living. This one has continued to bother me since I didn't actually learn about the states of matter in a classroom again until I took a thermodynamics course in college, which was when I learned about superfluids and superconductors requiring phase transitions as well. Superconductors and superfluids aren't really a part of everyday life, but plasmas are (the sun, florescent lights, etc).
1. Actually, my question was more intended as a probe into your target market (although re-reading it I feel it sounds rather rude. Sorry). I rarely uses hashes that need to process large amounts of data quickly. For work, I've used hashing to store passwords. For school/research, I had to hash a large list of keywords. For home use, I actually check my checksums on dowloaded stuff (sure I do).
So, I have single use small, many use small, and single use large. I am trying to imagine a scenario where I need a super high performance hash (well, I'll give you one: a commercial product comparable to tripwire). I imagine that your market would be people who hash lots of large blobs. If it is a single large chunk, then a slow hash is merely an inconvenience. If it is lots of small chunks, then IO probably dominates processing (in fact, when is this not the case?).
Have you encountered such a situation yourself? Is business good (as other poster commented, it appears as if you are just starting, so this is a positive-only litmus test)? Who is your target customer? Akamai maybe (any mirror-for-profit business)?
2. For reasons discussed in 1, I suggest you do encryption first. Fast encryption is always good.
1. Is that your sig or is that part of your comment? If it is part of your comment, please explain why it would give me a whole new view on performance. If it's your sig, then spooky how it was related to the topic.
2. Assuming your stuff is good, when are you going to code up SHA-1 (*MY* favorite hash)?
3. On the server side of things, I would argue that correctness is more important than otherwise. If an app crashes 1 in 100 times for a desktop user, the developer blames windows and the user is satisfied (don't flame me on this, please). On the server, if the app crashes 1 in 100 times, it may bring down the transactions for 100s of users, making things very bad for the developer. For non-crash correctness problems, consider a problem which makes a minor, but cumulative error in subsequent runs. That would likely be disasterous for the server situation.
As far as clarity, find me one developer who has taken over a project and not complained about the quality of the inherited code ever. Seriously. (that's not directed at parent)
Not that it isn't insightful, but it *is* funny too.
(awww crap, can't post AC because I am in the UCSB subnet and some of my classmates are jackasses who cause AC posting to be disabled. JERKS!... I guess I better add something other than an MPU request)
As pointed out by others, Microsoft does release their patches for these worms. People just aren't updating.
Beyond the first fault (releasing software with holes in the first place), what are they doing wrong? They are releasing patches. They are advertising the fact that the patches are important. They have changed their windows update page to be more informative (the 1,2,3's of making sure your computer doesn't rape and pillage the internet). What more should they do?
I don't disagree with you, but I don't think it is quite so black and white. Should you consider the number of IIS servers that are serving dynamic vs static content, and do the same for apache? Should you look at how many IIS boxes are online and serving NO content (quite a lot would be my hypothesis)? The point is that you have boxes that are needlessly exposing themselves to security holes for functionality that is not used.
There are a number of other factors to consider. Mac OS X has ssh, ftp, etc all disabled by default. What security holes exist in the default installation? What security holes exist in a "common" installation? I can't really guess for either Windows or Mac. But, for giggles, let's bring another OS into the discussion: Fans of... OpenBSD (?) speak proudly of not having a single vulnerability in the default install. How many users of that system stick with the default install? If the number is low, how secure is the average system?
For a system like windows, there exist(ed) many applications which install MSDE (sql server lite) with the default sa/blank user. This is what enabled slammer to be so effective (well, that and people actually installing SQL server without changing the sa password).
As a funny anecdote: In the middle of one of the big worm attacks, a buddy of mine wanted to patch his laptop. He accidentally plugged his firewall/router daisychain cable into his cablemodem instead of his uplink cable (making the firewall useless). Before he had downloaded his patches, his laptop was infected. Cox@Home started filtering traffic on those ports within a week because the problem was so bad.
Is it the default for Apple's Software Update tool to run once a week? Mine runs once a day, but only if I leave my computer on at night (I usually do that, plus I am behind a firewall which port forwards nothing to the Mac). However, I can't remember what the defaults were, so I don't know what happens for users who turn their computers off before heading to bed. If the default is to check at 2am and there is no fallback plan, there are probably quite a few Macs running without the latest patches.
What features in OOo are you using? I actually gave up on office a while ago (except at work, where my employer paid for the license). At home and at school, I work almost exclusively in OO. That said, I must admit that excel is superior to OO's spreadsheet tool. I frequently generate data to be graphed as line graphs or bar graphs. A line graph with 3 columns of 2000 data points takes slightly less than forever to generate in OO, and it generates very fast in excel. Similarly, When producing bar graphs, it is often convenient to have descriptive (read: longer than 4 characters) labels on the X-axis. This feature is horribly broken in OO. Try it. You have a choice of truncating long names (90 degree rotation doesn't extend the graph properly) or having the text print out in ugly vertical columns with horizontal lettering). It's as if the OO team never use their own graphing tools. (yeah, I know, stop bitching and pitch in and help...)
I think Open Office is a very good tool. I like the fact that it prints to pdf. Most of the interface is extremely easy to use. However, the product is not as polished as Office in many respects.
Lacking an access work-alike is also a detriment. Further, I am surprised they don't mention Project. I know many people who (unfortunately) think of their information in project as more important than information coming from the working team. "Project says we are half done! That means we'll be able to move our release up a week!" *shudder*
I admit that the advertising from microsoft criticizing that OO doesn't come with an email client is a bit off base--I would claim that not including Outlook is a security feature:).
You may feel different after using DB2. Since SQL is a terrible standard for stored procedures (IMHO), it is nice to have an embedded JVM for writing stored procedures in Java. After learning T-SQL (and having a short stint as a PL/SQL code monkey), I really have learned to hate proprietary SQL extensions. T-SQL isn't even all that proprietary--it looks a lot like SQL3. However, the implementation sucks. If I can't come up with a non-procedural implementation for a query (meaning I can't write it as a single query and need cursors), then it is faster to suck the data out of the database and loop over it using ADO than it is to use a cursor. That is terrible! Also, the dynamic capabilities of T-SQL suck. Want to assemble a statement based on user preferences? Too bad, string concatenation and EXEC are the only way to go (and they are limited and slow). I don't understand why using visual basic to cat strings together and submit them over the wire is faster than having a stored procedure do the same thing. Anyway, one of the features I was looking forward to in the next release of SQL server was just that, faster looping for procedural business logic. SQL Server 2000 already supports extended stored procedures, which have much of the same functionality, but with more overhead than I really want.
Yeah, I should have given more detail. Basically, my school / the federal government (FAFSA) required: 1) I had a job 2) I rented a place that was not my parents' address for 1 year (residency requirement for California) 3) Was more than 18, 4) was not claimed as a dependent on the previous year's income tax / claimed to be independent on my previous year's income tax, AND 5) had a letter from my parents saying I was independent.
I really liked the last requirement.
I didn't meet the requirements until my senior year of college. I ended up not qualifying for much.
But, in line with the parent poster: I had a friend who was independent (parents threw him out when he was 18 and basically cut all ties), but had a lot of trouble gaining independent status because his parents still claimed him as a dependent on their taxes. He went to a lawyer to try to get things fixed and was basically told that even if he sued his parents (for what, I'm not sure, but the damages were financial and easy to prove) he would likely still not qualify as an independent for a year. He was claiming himself as independent on his income tax and never got audited. Presumably his parents stopped claiming him as a dependent after he got too old. NOTE: tell this story at a college party with more than 2 people present and you will probably find someone in the same situation (at least at UCSB). MANY people get no support (financial or otherwise) once they leave for college and still don't qualify as independent. My parents were very supportive, they just tried to make me financially independent for my benefit (although they did change the locks the week after I left for school).
Now, to add another income avenue: When you start school, try very hard to find a professor willing to take on a Freshman as slave labor. Working in a lab (or doing research in general) is often a rewarding experience on its own--many professors will add your name to papers. It helps secure a good letter of recommendation for grants your university offers internally. Your department head, favorite professor, and undergraduate advisors may have information on free money with few applicants. Attend your professors' office hours and try to talk about their research (hint: read their recent publications first).
I second the small scholarship route. My girlfriend got a check every quarter from a local scholarship group. Check both local to your high school and local to the college you are attending.
Other options for funding:
1. As I recall, there is also the FAFSA (federal application for student aid, or something like that). When you get accepted into school in the US, the school may require you to fill one out. Basically, in the "white guy with average parents" scenario, you don't qualify for squat in grants. However, subsidized loans are better than bank loans.
2. If you want to plan better, look forward to year 2 at school. Start by getting a part time job and having your parents NOT claim you as a dependent on their taxes (this is easier if you are already 18). Once you become independent, your qualifications for government grants go way up (since you don't have your parents' income keeping you above the poverty line). The grants you can qualify more than make up for the tax credit. You can live like a king your freshman and sophmore years for less than 15K (total) income per year in California (note: a King in the freshman year of college has roommates and eats Ramen... royalty lives better outside of California). Of course, that assumes in state tuition. Tell your parents you will pay them back the tax credit for this year when you graduate. Hopefully they will support that idea as a zero interest loan or as a grant from Bank of Family.
3. Take every AP test you can, even if your school doesn't offer the corresponding course. If you pass 10 AP tests, there are a number of free scholarships you can get just by applying. It's quite possible too... if you are good at standardized tests.
4. Academic Scholarships: if you don't qualify for your freshman year, work your ass off and try for your sophmore year. 4.0 GPA = free money.
5. Psychology department: every school that has graduate students has a signup sheet for psych experiments which will pay 5-10 bucks. Except for my irrational fear of chalk, there's no downside to this. Basically, you get paid to breath for an hour.
6. Tutoring: this generally pays better than psych experiments (minimum wage pays better than psych experiments). Only slightly more work.
7. Parking Services, Library Security, Rec Center Front Desk: These are basically jobs where you get paid to study. DO NOT DO FOOD SERVICES!!!
8. If you aren't going to an in state school, try to get an apartment near your school to live. Then apply for residency your next year. That will lower your costs substantially. For some reason, the school will tell this to every graduate student but not to a single undergraduate. Again, this one sort of requires the part time job and financial (tax) independence.
Re:Does IT Matter?
on
Does IT Matter?
·
· Score: 2, Insightful
Please, read the article. He is saying that IT doesn't matter in the same way that the rail or electric industries don't matter. You pay your electric bills to keep the electricity flowing, but most companies should not be investing assloads of cash into, say, alternative energy sources. Instead, when an alternative energy source comes along that is better than what you've got, switch.
As a company, I do not need to invest heavily to get any of the advantages you list. They are a commodity... they don't matter. I am not going to invest a quarter of my R&D budget to improve my email service.
This isn't saying that there isn't a market for companies that DO invest heavily in IT. Certainly there are companies that invest heavily in improving other commodities. The point is that not every company needs to do this. During the 90's, there was the impression that you had to invest in building new technology, even if your business was making wicker baskets. The article argues that a basket making company should not invest heavily in IT. It should outsource IT and pay the dues it needs to keep its systems running.
OK, Now that I have had a rather lengthy RTFA, let me give my views. I agree that basket making companies should not invest to heavily in IT. However, I think that there are still enough aspects to IT that make IT still matter. Most businesses suffer from a lot of redundancy that could be automated, and that redundancy differs from business to business. In other words, there is an opportunity for that business to use its IT budget to save money, but that openning is not worth someone else building the infrastructure for it. Businesses must invest in themselves, and IT still promises to have one of the highest returns on investment for most businesses. Of course, the basket company should set that as a second priority to investing in baskets, but IT is still import--it can yeild a competitive advantage.
I may be feeding the IT doesn't matter argument--I am not as convinced that other companies wouldn't be able to be followers and acheive the automation cheaper than the company that did the initial investment. However, that is what intellectual property laws may prove beneficial for. Then companies still have an incentive to be the first. However, I am sure that at some point the patent office will get tired of pattent applications claiming "A device to automate X", where X is the business process of the month.
Of course, just about all of that undocumented stuff is pro-Linux
You might want to support this a little. As Microsoft cracks down on piracy your statement becomes more and more true (because there can be less undocumented installs of Windows--all sales are documented). However, I still know of quite a few people running cracked versions of XP or 2K. As such, I can't really estimate what percentage of undocumented stuff is pro-Linux. I have a red hat box at home. I downloaded the ISOs from a red hat ftp site. I _assume_ that is a documented install, but can't be sure. How many installs are not documented? Again, it is hard to estimate. Certainly people burning Linux distro CDs and sharing them with friends _might_ be undocumented, but the various vendors may already be assuming some level of sharing, which means that those numbers are documented. Anyway, it seems to me that as long as there is abundant Windows piracy, you can't really support your claim.
I am being nit-picky here, but I am also wondering if you have other data to support your claim (say, if you work at google or akamai and look at the logs to see all of the different machines that connect--since IE uses a non-standard HTTP protocol, you might have better estimates).
That's why we should throw a party at the event horizon. Everyone arrives at the same time and the party lasts forever.
That or nobody ever gets there and the ride is extremely short.
I can't remember which was the inside observer and which was the outside observer. I think it mixes reference points. The same time reference point is short, and the never arrive takes forever.
Yeah, but we also only BELIEVE that gravity will continue to function tomorrow. However, there is a wealth of evidence that it will. There is a wealth of evidence for black holes, as black holes in binary systems have been found (of course the black hole could just be a really dim, massive object, but it still is pretty convincing).
Radically Bent: Light does get bent around massive objects. The Einstein Cross and the Einstein Rings are predicted artifacts from viewing a glowing object behind a massive object. Google search on "Hubble deep field picture" and you will see some nice examples of this.
180 Bent: As far as the orbit being bent back at you, that would imply an eliptical orbit, which would have to cross inside the path of a circular orbit. I am far too lazy to solve for it right now, but I think that the light would end up having to travel through the event horizon for such an orbit. Still, there are some cool theory papers about bouncing lasers off of a spinning black hole (Kerr holes) and stealing the rotational energy of the black hole to pump up the laser energy and turn it into a superweapon. And "bounce" isn't like mirror, but more like a gravity whip for satellites.
One way of generating heat in the earth is through tidal forces. The earth is slowly being kneaded like a big doughball by the sun and the moon. This converts rotational energy into thermal energy, slowing the earth's spin down. A violent illustration of this effect is the Jovian moon IO (ok, this whole post was just to use the term Jovian). Io's volcanic activity is almost all generated heat in the core by tidal effects from Jupiter. Since the core is mostly thermally isolated the Earth really doesn't dissipate this energy well. Of course, the tidal forces for Earth are really weak and I have no idea what effect they have on core temperature, if any. But, take a look at Io in a decent astrophysics text.
Another possibility which has happened in the Earth's history is deposits of fissile material moving past one another due to geological plate movement and reaching critical mass. It takes millions if not billions of years, so it doesn't act like a nuke, but fission does occur and heat is created. One such occurance has been discovered on the African continent. I think there was an article in Nature about 4 years ago about it.
Anyway, two mechanisms for the Earth generating heat with no obvious external source (tidal forces come from the sun and the moon, but nothing as noticable as the sun going nova). So no, you aren't showing ignorance (both of these scenarios are less than impressive on the Earth), but there are possibilities (and they do obey Thermodynamics).
You may want to look for some hard core CS students to hire. I have a student who wrote a reverse compiler for microsoft.net (because he could) that outputs C# code. He also wrote a.Net compiler for brainf*ck (because he could). He basically views it all as source code transformations, and while I agree with him at a conceptual level, I do not have the patience nor his expertice at writing such code to make any recommendations to him. The guy can write a parser for a language faster than anyone I've ever met. And he's an undergraduate.
Wasn't ADA developed by the DoD for use in all DoD applications? There are still some legacy applications that use ADA, but I hope that you are right and it is dead. Can anybody who does embedded programming for DoD verify this?
Looking at ADA is painful. It looks like a language developed by a very large committee that wanted a language to do everything well. It acheives this, at the cost of being next to impossible to learn.
Thanks, I'll have to look into these. Sad Disclaimer: I am teaching a college course in Distributed Computing this Fall. I'm using Java as the implementation language (why? Because it's at a University and the students learn Java first). I was planning on using XML message passing because I have a student that refuses to work in Java (so he ports all of my framework to C++) and I am loath to use RMI. My bandwidth requirements are very low--reconfiguring my problem requires sending an objective function, the clients are genetic algorithm machines, so they consume memory without consuming bandwidth. The distributed computer is headless and unreliable (I am having students with dial-ups participate). My problems for my students all come from my real job working for a defense contractor (mostly scheduling, resource management, and similar OR problems). Due to academia's love for XML I will probably still use it, but I will probably suggest rewriting the protocol in one of the ones you just offered as a group project (and what better way to learn something than to assign it to undergraduates;) ?).
As far as the Hack mentality (of ignoring tags): labeling the schema of your XML messages is actually a fantastic low cost way of getting disperate systems to talk to one another. Your argument seems to assume that it is possible to upgrade both the client and the consumer at the same time (but if you've got a better way, I am all ears). In the Navy right now, there are several different contractors with products that talk to each other and have different funding cycles, so the ammount of data passed increases with each version of each product, independent if the listener has been upgraded--and given that we are often fighting for the same dollars, odds are one of us will be upgraded, then the next, then the next, etc. The extensibility model works really well for us in that environment. I put extra information in my XML output, when you get your funding, you actually parse and use them. However, if I have my way, I'll get my next round of funding before you get the chance.
Your point on toasted data is well made.
We use redundant RAID (yes, that's Redundant Redundant Arrays....) devices to store data, so toasting data due to device failure isn't really a concern. However, if you give them a chance, your customers' admins will shoot themselves in the foot and call DELETE with no WHERE clause and cry to you. If the system is functioning properly, all redundant live systems have the DELETE go through. Your colocated database is also trashed by user error. NOW, you still have the backups and the live transaction log that takes you back to your happy period right before the DELETE. In my experience, user error strikes much more frequently than expensive redundant hardware failure.
And just to throw off the "Why did your user have admin rights to the database?" At my current real job, my customer is the Navy, and up until recently, they had their own IT staff and didn't trust contractors (which for the most part is good policy) for admin purposes. This issue is even more compound now that the Navy has contracted out all of their IT duties and an employee of a competitor is hired to admin my databse system used by the Navy. (dreamed up example follows) Hmmm... no conflict of interest there: I win a bid for a contract. I build a system. I deliver it. Employee of company who lost the bid admins my system. OOPS! He "accidently" deleted data from my system. He blames poor design on my part (which sucks because I am required by contract to give him full admin rights). Perhaps his company should get the next contract to make such a system?
But I am paranoid...
Either way, I don't see how backups are not useful.
What do you suggest for distibuted computing if not XML messaging in some form? IMHO, there is really no way that a C++ developer is going to be able to communicate with a remote Java application except by passing XML. The Java RMI protocol is hard to implement in another language (it serializes exceptions and pipes them across to the remote client in a non-obvious way). I don't know enough about CORBAs model to comment, but opening a socket and sending strings (with inline XML schemas to verify against) is straighforward for all languages. Building your own parser for anything sucks, but there are XML parse libraries freely or cheaply available for any programming language. Also, XML messaging is convenient because you can upgrade a server to provide extra information / functionality without upgrading clients: just spec that unknown tags are parsed and ignored. Also, most XML parsers don't choke on malformed tags. If you write your own protocol using a home brew format, specing that kind of stuff is a real pain in the ass.
Slashdot Mirror
What mass/volume of this plant technology is required to recycle the CO2 for a single person? Also, does this technology have any consumables of its own such as H20? Seriously, do you think an office plant provides you with enough air to sustain your life?
As far as producing 02 and Water on Mars, how much equipment is required for that (and include the maintenence equipment)? Columbus was taking a risk that he would get to Asia before food ran out. He knew that he could buy his replentishment supplies in Asia. Furthermore, he knew that his trip back would take about the same time as his trip there (Asia/Europe didn't move closer/further as his trip progressed). He didn't plan on hitting a continent before Asia. What if he had hit a giant desert continent with no discernable food source? Ok, teach a man to fish... but the water problem remains unless they had decent destillers or something for getting drinking water from salt water. History is not my best subject.
By the way, it is going to take a lot of propellant to come back (Mars gravity is close to Earth gravity). That implies that you have to store it in those same discarded (oops) boosters that you took off of earth in (or replacements sent to mars on a seperate trip). We can't take off of earth effectively without a costly launch platform, what makes people assume we can do it on mars (because mars is all rocky, just like the moon so we can use the same launch technology)? How are we going to get that set up? Those air force pilots better bring along a couple of good constuction workers. Hell, just landing a craft upright will take a lot of fuel (or you need to build a big ass crane to upright it after it lands, which exacerbates the mass problem).
OK, now I am going to RTFA...
because she wants to know what her son is doing 18 hours a day tied to the computer
:).
And you can't think of anything better to replace "I was downloading porn" than "I was programming?"
Sheesh.
I can't help bust ask "why"?[sic]
Answering that question would provide insight as to what to teach her. If she just wants to understand the draw of programming, there are probably better ways to make her understand than to teach her programming
In what context is Visual Basic too difficult? If presented as a blank slate followed by "and here's how to write a compiler in VB" I can see the difficulty. The materials used for teaching matter a lot more than what is being taught. Even HTML is a pain in the ass for beginners without a proper tutorial: "which way does the slash go?", "is h2 bigger than h1?", "are you sure I can't break the computer?"
I remember a SAM's publishing book that was a carryover from Peter Norton's Guide to Visual Basic (They changed the name after VB2 to Visual Basic 3/4/5/6 in 21 Days/24 Hours, something like that). I have used various versions of that book to teach "yes you have to plug it in first" types the basics of programming. One of the first exercises is a simple alarm clock (using a timer control and the system beep). Then a simple text editor with copy and paste. After that, you should provide your own fractal drawing sample (a simple recursive function with lineto's).
Using the above reference, VB is easier to learn than just about anything else (including HTML). It is/was easily the best book published by Sam's, and probably one of the best programming tutorials as well (and I have the little Lisper, My First C/C++ program, and a collection of other tutorials that I haven't had as much luck with).
Incidently, even if you are using a self running tutorial, make sure you are sitting nearby and well versed in first aid. I've seen people come close to panic attacks when the IDE crashed on them.
Teachers fear technology replacing them, but they only have themselves to blame as they did not keep pace with it.
I had some good teachers (and some poor teachers) when I went through school. My parents are firm believers in the public school system, but they always took care to undo some of the misinformation that is taught in public schools. My faith in the public school system isn't nearly as strong as my parents'.
I spoke to a girl in a teaching credential program a few months ago. She said something to the effect of, "I would be a great teacher. I'm terrible at math. I'd tell them that they don't have to learn it because, 'hey, I turned out fine!'"
My mother is on the school board in my home town. She had to explain to teachers the concept of negative numbers (they were faced with budget cuts because they had a deficit). She had to make an analogy to credit cards before the teachers understood. Grant it, these were not math teachers and were at the k-3 school, but seriously, WTF?
Another story, the "math guru" teacher at the same school was at a school board meeting complaining about the uncompensated extra hours teachers put in. His example: "I have a high school algebra textbook that I keep by my bed that I sometimes read FOR FUN." Again, WTF? A teacher who dubs himself a "math guru" doesn't know algebra? I actually had this teacher when I was in 3rd grade for math (18 years ago). I did learn my multiplication tables from this guy.
I was a big geek in elementary school (ok, still am). I remember arguing with a science teacher because I knew that there were more states to matter than solid, liquid, and gas (such as plasma, superfluid, superconductor--I only knew of plasmas at the time). I actually had to argue that since a gas undergoes a second order phase transition (didn't know what that was, but I had been told about it), that it was a different state of matter when it became a plasma. I ended up getting sent outside for disrupting the class. All this because I asked my dad (a plasma physicist) what he did for a living. This one has continued to bother me since I didn't actually learn about the states of matter in a classroom again until I took a thermodynamics course in college, which was when I learned about superfluids and superconductors requiring phase transitions as well. Superconductors and superfluids aren't really a part of everyday life, but plasmas are (the sun, florescent lights, etc).
1. Actually, my question was more intended as a probe into your target market (although re-reading it I feel it sounds rather rude. Sorry). I rarely uses hashes that need to process large amounts of data quickly. For work, I've used hashing to store passwords. For school/research, I had to hash a large list of keywords. For home use, I actually check my checksums on dowloaded stuff (sure I do).
So, I have single use small, many use small, and single use large. I am trying to imagine a scenario where I need a super high performance hash (well, I'll give you one: a commercial product comparable to tripwire). I imagine that your market would be people who hash lots of large blobs. If it is a single large chunk, then a slow hash is merely an inconvenience. If it is lots of small chunks, then IO probably dominates processing (in fact, when is this not the case?).
Have you encountered such a situation yourself? Is business good (as other poster commented, it appears as if you are just starting, so this is a positive-only litmus test)? Who is your target customer? Akamai maybe (any mirror-for-profit business)?
2. For reasons discussed in 1, I suggest you do encryption first. Fast encryption is always good.
1. Is that your sig or is that part of your comment? If it is part of your comment, please explain why it would give me a whole new view on performance. If it's your sig, then spooky how it was related to the topic.
2. Assuming your stuff is good, when are you going to code up SHA-1 (*MY* favorite hash)?
3. On the server side of things, I would argue that correctness is more important than otherwise. If an app crashes 1 in 100 times for a desktop user, the developer blames windows and the user is satisfied (don't flame me on this, please). On the server, if the app crashes 1 in 100 times, it may bring down the transactions for 100s of users, making things very bad for the developer. For non-crash correctness problems, consider a problem which makes a minor, but cumulative error in subsequent runs. That would likely be disasterous for the server situation.
As far as clarity, find me one developer who has taken over a project and not complained about the quality of the inherited code ever. Seriously. (that's not directed at parent)
Not that it isn't insightful, but it *is* funny too.
... I guess I better add something other than an MPU request)
(awww crap, can't post AC because I am in the UCSB subnet and some of my classmates are jackasses who cause AC posting to be disabled. JERKS!
As pointed out by others, Microsoft does release their patches for these worms. People just aren't updating.
Beyond the first fault (releasing software with holes in the first place), what are they doing wrong? They are releasing patches. They are advertising the fact that the patches are important. They have changed their windows update page to be more informative (the 1,2,3's of making sure your computer doesn't rape and pillage the internet). What more should they do?
I don't disagree with you, but I don't think it is quite so black and white. Should you consider the number of IIS servers that are serving dynamic vs static content, and do the same for apache? Should you look at how many IIS boxes are online and serving NO content (quite a lot would be my hypothesis)? The point is that you have boxes that are needlessly exposing themselves to security holes for functionality that is not used.
... OpenBSD (?) speak proudly of not having a single vulnerability in the default install. How many users of that system stick with the default install? If the number is low, how secure is the average system?
There are a number of other factors to consider. Mac OS X has ssh, ftp, etc all disabled by default. What security holes exist in the default installation? What security holes exist in a "common" installation? I can't really guess for either Windows or Mac. But, for giggles, let's bring another OS into the discussion: Fans of
For a system like windows, there exist(ed) many applications which install MSDE (sql server lite) with the default sa/blank user. This is what enabled slammer to be so effective (well, that and people actually installing SQL server without changing the sa password).
As a funny anecdote: In the middle of one of the big worm attacks, a buddy of mine wanted to patch his laptop. He accidentally plugged his firewall/router daisychain cable into his cablemodem instead of his uplink cable (making the firewall useless). Before he had downloaded his patches, his laptop was infected. Cox@Home started filtering traffic on those ports within a week because the problem was so bad.
Is it the default for Apple's Software Update tool to run once a week? Mine runs once a day, but only if I leave my computer on at night (I usually do that, plus I am behind a firewall which port forwards nothing to the Mac). However, I can't remember what the defaults were, so I don't know what happens for users who turn their computers off before heading to bed. If the default is to check at 2am and there is no fallback plan, there are probably quite a few Macs running without the latest patches.
What features in OOo are you using? I actually gave up on office a while ago (except at work, where my employer paid for the license). At home and at school, I work almost exclusively in OO. That said, I must admit that excel is superior to OO's spreadsheet tool. I frequently generate data to be graphed as line graphs or bar graphs. A line graph with 3 columns of 2000 data points takes slightly less than forever to generate in OO, and it generates very fast in excel. Similarly, When producing bar graphs, it is often convenient to have descriptive (read: longer than 4 characters) labels on the X-axis. This feature is horribly broken in OO. Try it. You have a choice of truncating long names (90 degree rotation doesn't extend the graph properly) or having the text print out in ugly vertical columns with horizontal lettering). It's as if the OO team never use their own graphing tools. (yeah, I know, stop bitching and pitch in and help...)
:).
I think Open Office is a very good tool. I like the fact that it prints to pdf. Most of the interface is extremely easy to use. However, the product is not as polished as Office in many respects.
Lacking an access work-alike is also a detriment. Further, I am surprised they don't mention Project. I know many people who (unfortunately) think of their information in project as more important than information coming from the working team. "Project says we are half done! That means we'll be able to move our release up a week!" *shudder*
I admit that the advertising from microsoft criticizing that OO doesn't come with an email client is a bit off base--I would claim that not including Outlook is a security feature
You may feel different after using DB2. Since SQL is a terrible standard for stored procedures (IMHO), it is nice to have an embedded JVM for writing stored procedures in Java. After learning T-SQL (and having a short stint as a PL/SQL code monkey), I really have learned to hate proprietary SQL extensions. T-SQL isn't even all that proprietary--it looks a lot like SQL3. However, the implementation sucks. If I can't come up with a non-procedural implementation for a query (meaning I can't write it as a single query and need cursors), then it is faster to suck the data out of the database and loop over it using ADO than it is to use a cursor. That is terrible! Also, the dynamic capabilities of T-SQL suck. Want to assemble a statement based on user preferences? Too bad, string concatenation and EXEC are the only way to go (and they are limited and slow). I don't understand why using visual basic to cat strings together and submit them over the wire is faster than having a stored procedure do the same thing. Anyway, one of the features I was looking forward to in the next release of SQL server was just that, faster looping for procedural business logic. SQL Server 2000 already supports extended stored procedures, which have much of the same functionality, but with more overhead than I really want.
Microsoft research releases a lot of stuff that is portable. Also, I recently downloaded MSN Messenger client for my Mac (used to be able to get IE).
Yeah, I should have given more detail. Basically, my school / the federal government (FAFSA) required: 1) I had a job 2) I rented a place that was not my parents' address for 1 year (residency requirement for California) 3) Was more than 18, 4) was not claimed as a dependent on the previous year's income tax / claimed to be independent on my previous year's income tax, AND 5) had a letter from my parents saying I was independent.
I really liked the last requirement.
I didn't meet the requirements until my senior year of college. I ended up not qualifying for much.
But, in line with the parent poster: I had a friend who was independent (parents threw him out when he was 18 and basically cut all ties), but had a lot of trouble gaining independent status because his parents still claimed him as a dependent on their taxes. He went to a lawyer to try to get things fixed and was basically told that even if he sued his parents (for what, I'm not sure, but the damages were financial and easy to prove) he would likely still not qualify as an independent for a year. He was claiming himself as independent on his income tax and never got audited. Presumably his parents stopped claiming him as a dependent after he got too old. NOTE: tell this story at a college party with more than 2 people present and you will probably find someone in the same situation (at least at UCSB). MANY people get no support (financial or otherwise) once they leave for college and still don't qualify as independent. My parents were very supportive, they just tried to make me financially independent for my benefit (although they did change the locks the week after I left for school).
Now, to add another income avenue:
When you start school, try very hard to find a professor willing to take on a Freshman as slave labor. Working in a lab (or doing research in general) is often a rewarding experience on its own--many professors will add your name to papers. It helps secure a good letter of recommendation for grants your university offers internally. Your department head, favorite professor, and undergraduate advisors may have information on free money with few applicants. Attend your professors' office hours and try to talk about their research (hint: read their recent publications first).
I second the small scholarship route. My girlfriend got a check every quarter from a local scholarship group. Check both local to your high school and local to the college you are attending.
Other options for funding:
1. As I recall, there is also the FAFSA (federal application for student aid, or something like that). When you get accepted into school in the US, the school may require you to fill one out. Basically, in the "white guy with average parents" scenario, you don't qualify for squat in grants. However, subsidized loans are better than bank loans.
2. If you want to plan better, look forward to year 2 at school. Start by getting a part time job and having your parents NOT claim you as a dependent on their taxes (this is easier if you are already 18). Once you become independent, your qualifications for government grants go way up (since you don't have your parents' income keeping you above the poverty line). The grants you can qualify more than make up for the tax credit. You can live like a king your freshman and sophmore years for less than 15K (total) income per year in California (note: a King in the freshman year of college has roommates and eats Ramen... royalty lives better outside of California). Of course, that assumes in state tuition. Tell your parents you will pay them back the tax credit for this year when you graduate. Hopefully they will support that idea as a zero interest loan or as a grant from Bank of Family.
3. Take every AP test you can, even if your school doesn't offer the corresponding course. If you pass 10 AP tests, there are a number of free scholarships you can get just by applying. It's quite possible too... if you are good at standardized tests.
4. Academic Scholarships: if you don't qualify for your freshman year, work your ass off and try for your sophmore year. 4.0 GPA = free money.
5. Psychology department: every school that has graduate students has a signup sheet for psych experiments which will pay 5-10 bucks. Except for my irrational fear of chalk, there's no downside to this. Basically, you get paid to breath for an hour.
6. Tutoring: this generally pays better than psych experiments (minimum wage pays better than psych experiments). Only slightly more work.
7. Parking Services, Library Security, Rec Center Front Desk: These are basically jobs where you get paid to study. DO NOT DO FOOD SERVICES!!!
8. If you aren't going to an in state school, try to get an apartment near your school to live. Then apply for residency your next year. That will lower your costs substantially. For some reason, the school will tell this to every graduate student but not to a single undergraduate. Again, this one sort of requires the part time job and financial (tax) independence.
Please, read the article. He is saying that IT doesn't matter in the same way that the rail or electric industries don't matter. You pay your electric bills to keep the electricity flowing, but most companies should not be investing assloads of cash into, say, alternative energy sources. Instead, when an alternative energy source comes along that is better than what you've got, switch.
As a company, I do not need to invest heavily to get any of the advantages you list. They are a commodity... they don't matter. I am not going to invest a quarter of my R&D budget to improve my email service.
This isn't saying that there isn't a market for companies that DO invest heavily in IT. Certainly there are companies that invest heavily in improving other commodities. The point is that not every company needs to do this. During the 90's, there was the impression that you had to invest in building new technology, even if your business was making wicker baskets. The article argues that a basket making company should not invest heavily in IT. It should outsource IT and pay the dues it needs to keep its systems running.
OK, Now that I have had a rather lengthy RTFA, let me give my views. I agree that basket making companies should not invest to heavily in IT. However, I think that there are still enough aspects to IT that make IT still matter. Most businesses suffer from a lot of redundancy that could be automated, and that redundancy differs from business to business. In other words, there is an opportunity for that business to use its IT budget to save money, but that openning is not worth someone else building the infrastructure for it. Businesses must invest in themselves, and IT still promises to have one of the highest returns on investment for most businesses. Of course, the basket company should set that as a second priority to investing in baskets, but IT is still import--it can yeild a competitive advantage.
I may be feeding the IT doesn't matter argument--I am not as convinced that other companies wouldn't be able to be followers and acheive the automation cheaper than the company that did the initial investment. However, that is what intellectual property laws may prove beneficial for. Then companies still have an incentive to be the first. However, I am sure that at some point the patent office will get tired of pattent applications claiming "A device to automate X", where X is the business process of the month.
Of course, just about all of that undocumented stuff is pro-Linux
You might want to support this a little. As Microsoft cracks down on piracy your statement becomes more and more true (because there can be less undocumented installs of Windows--all sales are documented). However, I still know of quite a few people running cracked versions of XP or 2K. As such, I can't really estimate what percentage of undocumented stuff is pro-Linux. I have a red hat box at home. I downloaded the ISOs from a red hat ftp site. I _assume_ that is a documented install, but can't be sure. How many installs are not documented? Again, it is hard to estimate. Certainly people burning Linux distro CDs and sharing them with friends _might_ be undocumented, but the various vendors may already be assuming some level of sharing, which means that those numbers are documented. Anyway, it seems to me that as long as there is abundant Windows piracy, you can't really support your claim.
I am being nit-picky here, but I am also wondering if you have other data to support your claim (say, if you work at google or akamai and look at the logs to see all of the different machines that connect--since IE uses a non-standard HTTP protocol, you might have better estimates).
Step 1.5: Bring donuts into the office. Don't bring napkins.
Then you can basically skip the chalk steps because you can identify the donut glaze.
The other method:
Step 1: Talk to the VP on the way to his office.
Step 2: Watch which buttons he presses and remember the code.
Simple solutions work with high frequency.
That's why we should throw a party at the event horizon. Everyone arrives at the same time and the party lasts forever.
That or nobody ever gets there and the ride is extremely short.
I can't remember which was the inside observer and which was the outside observer. I think it mixes reference points. The same time reference point is short, and the never arrive takes forever.
Isn't relativity fun?
Yeah, but we also only BELIEVE that gravity will continue to function tomorrow. However, there is a wealth of evidence that it will. There is a wealth of evidence for black holes, as black holes in binary systems have been found (of course the black hole could just be a really dim, massive object, but it still is pretty convincing).
Radically Bent:
Light does get bent around massive objects. The Einstein Cross and the Einstein Rings are predicted artifacts from viewing a glowing object behind a massive object. Google search on "Hubble deep field picture" and you will see some nice examples of this.
180 Bent:
As far as the orbit being bent back at you, that would imply an eliptical orbit, which would have to cross inside the path of a circular orbit. I am far too lazy to solve for it right now, but I think that the light would end up having to travel through the event horizon for such an orbit. Still, there are some cool theory papers about bouncing lasers off of a spinning black hole (Kerr holes) and stealing the rotational energy of the black hole to pump up the laser energy and turn it into a superweapon. And "bounce" isn't like mirror, but more like a gravity whip for satellites.
One way of generating heat in the earth is through tidal forces. The earth is slowly being kneaded like a big doughball by the sun and the moon. This converts rotational energy into thermal energy, slowing the earth's spin down. A violent illustration of this effect is the Jovian moon IO (ok, this whole post was just to use the term Jovian). Io's volcanic activity is almost all generated heat in the core by tidal effects from Jupiter. Since the core is mostly thermally isolated the Earth really doesn't dissipate this energy well. Of course, the tidal forces for Earth are really weak and I have no idea what effect they have on core temperature, if any. But, take a look at Io in a decent astrophysics text.
Another possibility which has happened in the Earth's history is deposits of fissile material moving past one another due to geological plate movement and reaching critical mass. It takes millions if not billions of years, so it doesn't act like a nuke, but fission does occur and heat is created. One such occurance has been discovered on the African continent. I think there was an article in Nature about 4 years ago about it.
Anyway, two mechanisms for the Earth generating heat with no obvious external source (tidal forces come from the sun and the moon, but nothing as noticable as the sun going nova). So no, you aren't showing ignorance (both of these scenarios are less than impressive on the Earth), but there are possibilities (and they do obey Thermodynamics).
Cheers,
Chris
Don't give up on the code.
.net (because he could) that outputs C# code. He also wrote a .Net compiler for brainf*ck (because he could). He basically views it all as source code transformations, and while I agree with him at a conceptual level, I do not have the patience nor his expertice at writing such code to make any recommendations to him. The guy can write a parser for a language faster than anyone I've ever met. And he's an undergraduate.
You may want to look for some hard core CS students to hire. I have a student who wrote a reverse compiler for microsoft
Wasn't ADA developed by the DoD for use in all DoD applications? There are still some legacy applications that use ADA, but I hope that you are right and it is dead. Can anybody who does embedded programming for DoD verify this?
Looking at ADA is painful. It looks like a language developed by a very large committee that wanted a language to do everything well. It acheives this, at the cost of being next to impossible to learn.
Thanks, I'll have to look into these. Sad Disclaimer: I am teaching a college course in Distributed Computing this Fall. I'm using Java as the implementation language (why? Because it's at a University and the students learn Java first). I was planning on using XML message passing because I have a student that refuses to work in Java (so he ports all of my framework to C++) and I am loath to use RMI. My bandwidth requirements are very low--reconfiguring my problem requires sending an objective function, the clients are genetic algorithm machines, so they consume memory without consuming bandwidth. The distributed computer is headless and unreliable (I am having students with dial-ups participate). My problems for my students all come from my real job working for a defense contractor (mostly scheduling, resource management, and similar OR problems). Due to academia's love for XML I will probably still use it, but I will probably suggest rewriting the protocol in one of the ones you just offered as a group project (and what better way to learn something than to assign it to undergraduates ;) ?).
As far as the Hack mentality (of ignoring tags): labeling the schema of your XML messages is actually a fantastic low cost way of getting disperate systems to talk to one another. Your argument seems to assume that it is possible to upgrade both the client and the consumer at the same time (but if you've got a better way, I am all ears). In the Navy right now, there are several different contractors with products that talk to each other and have different funding cycles, so the ammount of data passed increases with each version of each product, independent if the listener has been upgraded--and given that we are often fighting for the same dollars, odds are one of us will be upgraded, then the next, then the next, etc. The extensibility model works really well for us in that environment. I put extra information in my XML output, when you get your funding, you actually parse and use them. However, if I have my way, I'll get my next round of funding before you get the chance.
Your point on toasted data is well made.
We use redundant RAID (yes, that's Redundant Redundant Arrays....) devices to store data, so toasting data due to device failure isn't really a concern. However, if you give them a chance, your customers' admins will shoot themselves in the foot and call DELETE with no WHERE clause and cry to you. If the system is functioning properly, all redundant live systems have the DELETE go through. Your colocated database is also trashed by user error. NOW, you still have the backups and the live transaction log that takes you back to your happy period right before the DELETE. In my experience, user error strikes much more frequently than expensive redundant hardware failure.
And just to throw off the "Why did your user have admin rights to the database?" At my current real job, my customer is the Navy, and up until recently, they had their own IT staff and didn't trust contractors (which for the most part is good policy) for admin purposes. This issue is even more compound now that the Navy has contracted out all of their IT duties and an employee of a competitor is hired to admin my databse system used by the Navy. (dreamed up example follows) Hmmm... no conflict of interest there: I win a bid for a contract. I build a system. I deliver it. Employee of company who lost the bid admins my system. OOPS! He "accidently" deleted data from my system. He blames poor design on my part (which sucks because I am required by contract to give him full admin rights). Perhaps his company should get the next contract to make such a system?
But I am paranoid...
Either way, I don't see how backups are not useful.
What do you suggest for distibuted computing if not XML messaging in some form? IMHO, there is really no way that a C++ developer is going to be able to communicate with a remote Java application except by passing XML. The Java RMI protocol is hard to implement in another language (it serializes exceptions and pipes them across to the remote client in a non-obvious way). I don't know enough about CORBAs model to comment, but opening a socket and sending strings (with inline XML schemas to verify against) is straighforward for all languages. Building your own parser for anything sucks, but there are XML parse libraries freely or cheaply available for any programming language. Also, XML messaging is convenient because you can upgrade a server to provide extra information / functionality without upgrading clients: just spec that unknown tags are parsed and ignored. Also, most XML parsers don't choke on malformed tags. If you write your own protocol using a home brew format, specing that kind of stuff is a real pain in the ass.