And if you think IT is bad (I'm in data / VOIP networks), wait till you see how the business sees voice expenses (used to maintain PABXs), then compare to how loudly they scream when their phones (or even telephone banking 1300 IVRs etc.... I'm talking about banks here!) go down
As a CCNA and certified ericsson PABX tech with a liberal arts degree (honours in pol sci actually!!!) I should take offence
though as a slashdot reader I would rather laugh along with you through gritted teeth
arts degrees teach you some valuable skills missing in some vocational courses. If said CIO has enough experience in the field then what degree he/her has is irrelevant.
I would rather follow a cio who has a liberal arts degree, then couldn't find a good paying lobbying job so went helpdesk, level 2 engineer, level 3 engineer, manager, project manager, dept manager, CIO than an IT graduate who went straight into management
I think the environment has changed to the point where the 'tipping point' you describe above will be a lot harder to achieve. Back in the mid 90s not every office worker had their own computer, email account and NETWORK ACCESS (nevermind internet). If they had a computer it was not uncommon for it to be used as a glorified typewriter and calculator. Also it was normal NOT to have a computer at home.
Now everyone has a windows PC and lives off IE/word/excel/outlook. The barrier to entry so to speak has been raised many bars higher. How many non geeks do you know who would be willing to learn a new OS? Heck people don't even want to use firefox as they're too used to cllicking on the IE icon.
Also back in the mid 90s, if they wanted to build a home domain/network environment, most geeks had few other options except for NT unless they could afford a unix server at home. And NT was poor. Now the obvious option is linux, which is free, very heavily documented, highly reliable and can be made to do everything bar make coffee (pretty sure someone's getting around to that too!)
Hence IMHO you are kinda right if you substitute mac for linux - what you describe above has been one of the big reasons for linux taking over the server space. But I still disagree on the desktop side.
Also like the below poster mentions, linux is the weapon of choice amongst the geek elite.
I'm not suggesting for a moment that macs can't regain their dominant status (market share wise) but I don't think it will primarily be driven by the top down method you describe.
disclaimer: typing this from my macbook, with a fedora box as samba/squid/torrentflux+apache/ssh backend server, and seriously thinking about a hackintosh when I build my next desktop
i hate MS as much as the next linux running CLI geek but you make some great points.
I work for a telco that is despised by most people here (australia... no prizes for guessing which one LOL) who then outsource me to an oil company (I'm actually on the network management team as a cisco tech) so I can sympathize with amy MS employees wrongly slandered by all those holier than thou smarmy replies.
I'm in no position to comment on the subject matter, not being a neuroscientist or whatnot
But GODDAMN the writers' prose reads like a pompous 3rd rate copy of Neal Stephenson - talk about rambling. Woohoo you lost concentration whilst driving. And if you stretch that basic point out into 500 turgid words, throw in a wounded girlfriend etc. then you apparently have an article.
No I am not impressed with your descriptive prose or what you think of cameraphones. Reading post-modernist theory into 'where do you want to go today' may get you an Arts thesis but its still a load of wanky BS.
YES WE GET THE PICTURE, GET TO THE GODDAMNED POINT
Disclaimer: - For the record, I am a big Neal Stephenson fan, but the article just reminds me of his prose style without any of his substance - I also have an arts degree, I know first hand the kind of drivel you are forced to read and write
Finder is a PITA but why stuff with the Mac environment in such a major way for very little benefit (ideology aside). Konqueror > Finder is pretty much the only (FOSS arguments aside) reason I can think of for preferring KDE. Plus you KNOW there will be bugs, half your OSX native software won't work properly, etc.
Disclaimer: I run KDE on my desktop (fedora), my laptop is a macbook so I am familiar with both environments.
Well right now in Oz, the apple online store is still listing ATI 2600 Pros against trhe 4000 dollar AUD Mac Pro.
Appreciate your comment on the server RAM.
However if we're going server, where are my hotswappable scsi hard drives with (proper) hardware RAID???
Anyhoo I think apple really needs to release a power-user desktop box (i.e. sans monitor). I love my macbook but no way I am getting an imac to replace my desktop - I don't want (slower and more expensive) laptop components hard wired to a screen i can't replace or change out i.e. an IMAC.
If there was one that was ~$1600-$1700USD with quad core CPU, 2 gigs, > 500 Gig and an 8800GT that would be perfect. Considering you could build that config with ~$1200-1300USD I think that accurately reflects the premium / OSX tax.
I love my macbook and apple applications but I am not getting a Mac Pro with its complete overkill specs / prices. And what is with the ridiculous price for upgrades from the Apple shop? 700 bucks AUD for extra 2 gigs more RAM here in Australia, which is ~600USD. You can find 2 gigs RAM easily from any third party supplier / shop for ~120AUD. Also where are the high end Nvidia cards?!?! you're spending over 3k AUD and all you get is a freakin 8600GT (which is what $150 AUD value?!??!). So it will be awhile before my windows/linux Core2Duo rig is replaced my a mac.....
If its using legacy coax cable, it must remain a bus topology i.e. everyone shares the same bandwidth. Unless the protocol includes some new way of muxing each individual channel
I guess you've never dealt with people certified in India? My boss is Fijian Indian and he tells me point blank that he doesn't trust certifications from India. When I say 'fake' I guess I mean more people who have done the cert but have absolutely nous. Put them in front of a production system with incomplete documentation and a cabling disaster and watch them bomb. So I aguess I am agreeing with you with your 'book' comment.
My current shop is actually really good, the other engineers are old timers who can tell you the pin-out on a V35 connector or how X25 works etc and there is a wealth of experience. We also get free reign across not only the data network but the VOIP network as well. The best part is that we built our CCM cluster up ourselves from scratch (as opposed to somone like Dimension Data coming in over the top and plonking it in) together with corresponding network redesign (routing the new subnets, implementing across the board QoS, firewalls etc.) so everybody actually had to learn it from scratch and we all got sent to the nice CCVP classes.
I am speaking from former experience + general observations around the industry + what I see from my parent company (my current shop is outsourced shop, with mamy people from different outfits, its complicated) whose core business is actually PABX, the number of duff PABX 'experts' I have had to deal with is astounding, as well as PABX guys who claim to know VOIP but couldn't configure a simple Layer 2 switch properly let alone a router or a voice gateway etc.
Mate you can have multiple VMware hosts that transparently load share. Take one box down, everything (live OSs and Apps!) is transferred across to the remaining members of the cluster, you can work on that box in peace and nobody is any the wiser.
Having say 4 ESXs all running at 25-30$ gives you potentially four separate boxes for each virtualised app. With the transparent live load sharing I've mentioned above. Sorry not a server guy so scarce on details but I have talked to enough server guys to know its very real and very live in a lot of high availability environments (banks, oil companies).
If you make the mistake of putting redundant virtual hosts on the same real host then that's just bad design.
Good points you've raised. My solution would be to weed out the server guys with no social skills and promote the desktop guys with brains. Not easy to find either in some environments I know....
The issue is that there are so many cisco tards makes them obvious targets (compared to relatively few PABX guys). Also HR are idiots who equate CCNA with real world competence. + the endless stream of fake CCNAs (we all know about this one) or even worse, genuine but truly 'book' CCNAs with no real world ability, just can memorise theory. There are NO PABX EQUIVALENTS, you don't see anybody walking around with the CCNP equivalent for Lucent Definity or whatever who has only passed exams and never worked on production system.
In terms of cross discipline competence its 'six of one or half a dozen of the other' in my opinion.
I have met many PABX guys who's data knowledge is so poor they really shouldn't claim any knowledge at all. I don't know too many PABX guys who can configure inter-VLAN trunking and route between VLANs (something a real CCNA with real enterprise class experience is going to do in their sleep). Data knowledge cuts a lot deeper than just knowing how to subnet. Can you understand a complicated data flow e.g comes in from the public WAN into a DMZ thereby being NATted twice by two different firewalls? secure tunnels for remote cluster syncing? How about QoS? Yes auto qos works like a treat a lot of the time but especially on your WAN you are going to want to manually tune your Low Latency Queuing, implement RTP header compression, interleaving etc.
On the other hand, the cisco guys who are willing to learn (as opposed to the tards, which there are many more of than PABX guys alone so they're just more visible) have a much easier curve because understanding the voice environment is conceptual. User requirements, IVR flows, number range management - how is this so different from managing IP allocations - etc. Their data knowledge makes it easier to understand VOIP configuration e.g. gateway configuration, CCM configuration (plus VOIP depends on your data network for its 'backplane' as I like to think of it).
You can be a great PABX guy who also knows the CCM interface and manages their voice environment very efficiently - what do you do when you have issues with your voice gateways - and all you know how to do is reset the router or replace it? Or a site's WAN link is flapping leading to SRST getting mixed up? QoS issues - even diagnosing it is a QoS issue? How about when global security comes in with their new security regs so you have to put the callmanager behind a firewall?
Point being that you need to understand BOTH to do a proper job in VOIP.
I am a former PABX man (Ericsson MD110!!! Great boxes) and now I am a cisco tard (CCNA) in a team running a call manager environment so I think I'm well placed to call it from a semi objective perspective.
Sounds like putting the cart before the horse. Building a solution, then looking for the problem to use it on....
Its just another app hosted on another server in the server farm, people fire up their IE (yes I hate IE tied stuff as much as the next geek) and get a web based app. Which apparently, they didn't want in the first place.....
Also people not in IT do not understand that everything is tied together a certain way for reasons (some good, some bad) that are often outside of ITs control. The App does not get implemented in isolation. For example, if this said app would require changing user permissions in their Active Directory accounts (say to allow ActiveX scripts), firewall / ACL changes on the LAN etc. then it becomes a security issue first and foremost (as well as a royal pain in the butt). The bureaucratic fun begins, but all because of a very good reason (security).
Now if its just people being obstinate and the app was really a better solution (in the long run, taking into account cost of development, implementation, support, business continuity, training, disruption to existing operations), then kudos to him for long term vision.
However I'm with the cynics on this one. Mind you this is from a networking guy's perspective, what do I know of web development?
If you think Telstra's bad..... you ain't seen the likes of Soul, Optus, AAPT in action at enterprise level.
OPTUS:
I have had carrier ports (Onramp 30s) where we've had to abort TWICE IN A ROW because Optus couldn't deliver the service as requested (ie tagged and looped back at the krone frame). a.) they left it in the fibre cabinet, didn't even run a cable to the comms room the PABX was in, b.) they told us they sent someone again but they actually didn't, so we sent our guy back... who reported that everything was untouched..... c.)
AAPT:
When I did a mercifully brief stint at AAPT, I had an incident where someone in provisioning arbritarily cancelled 12 business grade ADSL services belonging to one customer. They spotted a billing issue and decided to cancel the services, despite the fact that you could clearly see regular traffic on links BELONGING TO AN ENTERPRISE CUSTOMER WITH OVER 50 SITES WITH YOU. Without even looking at traffic stats or asking ANY technical area. Even funnier, Telstra reassigned those ports in the local DSLAMs so some of the sites took TWO MONTHS to get their connection back.
SOUL:
Now I'm dealing with a carrier called Soul. This is a company who takes THREE BITES OF THE CHERRY to implement BGP routing with our main data centre. THEN, they tell us we need ANOTHER PLANNED CHANGE because the secondary link we ordered off them SPECIFICALLY TO PROVIDE REDUNDANCY - was on the same router at their POP. THEN, we discover through auditing, they had duplicated a subnet in the fallback routes for two separate sites (ie if both sites went to their fallback links at the same time, there would be an addressing clash in the carrier cloud). Hilarious.
Another story about Soul, there are cases where they CANNOT upgrade the IOS on their routers to resolve issues because they refuse to PAY MAINTENANCE ON THEIR ROUTERS. This is a carrier whose taking out ads everywhere, but they won't pay for support on their hardware.
Say what you will, and I know that Telstra's consumer grade support is as bad as anyone else's, but at the enterprise level, everybody in Australian telecoms will tell you that you get what you pay for, and if its business critical, go Telstra.
Try running VOIP over a data network with 600ms latency from end to end................... (not that I have, but I can imagine!!!!! LOL)
In the data environment we have to jump through all sorts of (non technical) hoops to get Quality of Service properly implemented end to end just to get wired VOIP running properly, I suspect wireless VOIP - without the ability to properly implement queueing and classification / policy mapping (sorry, Cisco-centric view, but as I said I'm a data guy:) ) from end to end, I have no idea how it 'just works'
Actually, I have a Nokia N73 that does my ipod/cell phone/PDA functionality all fine and dandy. Plus:
It also runs Tom Tom so quadruples as a GPS unit. It also handles 3G. It also syncs nicely with outlook via bluetooth. Its also got a 3MP camera.
+ it takes a horde of 3rd party symbian apps (some great, some not so good...)
Point being: OK the UI ain't as flash, but I'm used to it now and it doesn't slow me down too much. Sure touch screen would be nice but I'm not taking a cute interface over real functionality.
Since carrying it around I've never looked back, sure it ain't as great an mp3 player as a dedicated player, the camera pics are distinctly iffy compared to a real camera (but blows chunks out of most camera phones mind you, and perfectly adequate for happy snaps). This is the target Apple should be aiming for. There are lots of phones out there that are quite decent MP3 players when u throw in a memory card.
Of course, I'm missing the point as this will not matter one iota to the average non tech user but what the hey, here's one geek whose not buying the hype.
I'll agree some of the Enterprise stuff M$ has been coming out with is of a much higher quality than their previous efforts (Server 2003 vs NT for example!!! the entire AD/Exchange thingy that underpings 90% of corporate desktop infrastructure).
But mate leave the personal comments out of it. If you didn't understand the rest of his comments, fine, but no need to get personal. I take it you regard yourself as a top developer, well I'm not a developer but I don't see anything M$ is doing that is in any way cutting edge, some of it is very good but its not 'cutting edge' in the sense that its original and new. Doesn't mean its not any good, or that it may be the best tool for the job, but as for innovation, you gotta be kidding. (unless your idea of innovation is to completely bork existing open standards and implement your own then leverage your market share to make it the de facto standard, in spite of its suckiness).
And I'll disagree with your provincial view remark, a large chunk of the IT industry hates M$ aside from the wintel boys and they only tolerate it because it earns them a crust. Heck I'm in network services (ie a cisco shop) and everybody rags on windows.
I guess the previous poster stepped on a few nerves?
And how is running links over microwave news??????????
In my old job we've had microwave E1 links (standard 2 meg) linking CBD sites for years!!!
OK its not fantastic in terms of bandwidth etc. and specifically speaking, its used for voice channels, but you can send anything over the layer 2???? How is this news????
Then get a tech support job (w/ a good company) and move on up! Just because you have a degree doesn't mean you're entitled to start a few rungs up the ladder. It helps, but its not a magic promotion bullet.
Speaking as somebody who worked their way up from the helpdesk (my degree is in pol sci, go figure), I personally would hire someone w/ 2 years experience (say) over a graduate anyday. That is industry reality. Ditto for industry certs say CCNA (which is only 1 year part time) over a degree.
Unfortunatley this is not properly explained to uni entrants, not your fault I admit.
It also doesn't help that a lot of CS degrees (and I'm grossly generalising here, so no flames pls:) ) churn out grads who can do a half-a$$ed job at a variety of things. Unfortunately most jobs are better served by people who can do a good job at a much more narrow scope. E.g. the department supports the IP network - why hire a CS grad who knows a bit about IP and has logged onto a few routers in test labs, vs a guy who's spent a year or two actually working with them!!! The CS grad may also know java programming, visual C, system design, database design, etc. but its all irrelvant.
Sorry for the rant but that's just how I see things, no offence intended. Just bite the bullet and get a tech support job in a good field (i.e. stay the ---- clear of consumer support for telcos etc., go business support) and see where u go from there,.
point acknowledged, y'know i wasn't really disagreeing with most of your statements
Well mate ur not the only techie who also plays instruments. I'm a hip hop scratch nerd, and you know how us deck-heads are w/ tech. Before that I spun @ drum and bass parties (that was before I sold off all my d&b twelves to concentrate on hip hop) - no serrato scratch or final scratch etc., just good old fashioned 12"s all the way. Before that I was a classical pianist, got to grade 8 before I realised it was boring as **** to play. So yes I know where you're coming from re: music and tech working FOR the music.
What did I do? Well when you're 5000 miles from home (on assignment for THE MAN so to speak) you're deprived of your usual hobbies. Since its Diwali (festival of lights) over here in India, got together with a few other people over here and we set off some firecrackers. And got smashed. Killing time till I get outta here... after 9 weeks, you kinda get over the 'ooh I'm in India, how exciting' thingy:) And one of the first things I'm going to do when I get home is... buy a Mac (ahahahaha), no seriously, one of them nice new Roland MPCs u'know the ones you can plug a monitor into etc.? they're pretty sick eh.
not trying to start some kinda argument, but please go easy on the assumptions - ur not the only techie who also plays music.
cheers
Please please please no more Mac vs Linux vs Windows articles. We all know the score
- In corporate enterprise space, its MS hell, so deal with it - I would love to see mass linux migrations but just thinking of the scale of the task... (shudders) - Run linux on servers wherever possible + your IT shop has the expertise to maintain it - Graphic designers and music producers are the only people who can really benefit from a Mac in the workplace - On your personal / home / media PC, whatever the **** you want, who cares - If in doubt, run Windows (not talking about mission critical or specialist e.g. large scale batch processing etc. apps!!!), it will get the job done somehow
the cult of Apple is no better than Windows FUD, its just that (since the late 90s anyway) they actually have a superior product. if anything the vendor-lockin is even worse - OPEN STANDARDS PLEASE, I don't need a solutions provider (ie Apple), I want options so I can figure out my OWN solutions (ie my needs are that of a geek, not the average 'i want it to just work' user)
*Disclaimer* I use linux whenever possible i.e. I use windows when I have to (work) or I'm too lazy / linux-n00b to figure it out (media centre)
Been using Fedora for the last year or so, FC4 --> FC5, pretty happy in general. But as I read more and more about Slack (and run into more and more redhat/fedora issues) I start wondering: is it worth switching? I'll qualify the question (yes horses for courses etc. etc...:) )
-Usage: General PC stuff, lots of downloading, a bit of multimedia (mostly music, previewing videos I DL:) ) file server / SSH gateway. I would turn it into a full blown router but don't see the point of buying 3 more NICs when my little linksys box does the job happily, aside from having full iptables control
-Nix knowledge: somewhere between advanced newbie and average user. Only started using Nix 1-2 years ago but I like tinkering and am not scared of CLI:)
-Time: I used to devote a lot of time to geeking around with my box, but lately (esp. with work ramping up) I haven't been ars3d. Though a fresh challenge may increase my motivation!
Slashdot Mirror
Hear hear
And if you think IT is bad (I'm in data / VOIP networks), wait till you see how the business sees voice expenses (used to maintain PABXs), then compare to how loudly they scream when their phones (or even telephone banking 1300 IVRs etc.... I'm talking about banks here!) go down
As a CCNA and certified ericsson PABX tech with a liberal arts degree (honours in pol sci actually!!!) I should take offence
though as a slashdot reader I would rather laugh along with you through gritted teeth
arts degrees teach you some valuable skills missing in some vocational courses. If said CIO has enough experience in the field then what degree he/her has is irrelevant.
I would rather follow a cio who has a liberal arts degree, then couldn't find a good paying lobbying job so went helpdesk, level 2 engineer, level 3 engineer, manager, project manager, dept manager, CIO than an IT graduate who went straight into management
Sorry, I respectfully disagree.
I think the environment has changed to the point where the 'tipping point' you describe above will be a lot harder to achieve. Back in the mid 90s not every office worker had their own computer, email account and NETWORK ACCESS (nevermind internet). If they had a computer it was not uncommon for it to be used as a glorified typewriter and calculator. Also it was normal NOT to have a computer at home.
Now everyone has a windows PC and lives off IE/word/excel/outlook. The barrier to entry so to speak has been raised many bars higher. How many non geeks do you know who would be willing to learn a new OS? Heck people don't even want to use firefox as they're too used to cllicking on the IE icon.
Also back in the mid 90s, if they wanted to build a home domain/network environment, most geeks had few other options except for NT unless they could afford a unix server at home. And NT was poor. Now the obvious option is linux, which is free, very heavily documented, highly reliable and can be made to do everything bar make coffee (pretty sure someone's getting around to that too!)
Hence IMHO you are kinda right if you substitute mac for linux - what you describe above has been one of the big reasons for linux taking over the server space. But I still disagree on the desktop side.
Also like the below poster mentions, linux is the weapon of choice amongst the geek elite.
I'm not suggesting for a moment that macs can't regain their dominant status (market share wise) but I don't think it will primarily be driven by the top down method you describe.
disclaimer: typing this from my macbook, with a fedora box as samba/squid/torrentflux+apache/ssh backend server, and seriously thinking about a hackintosh when I build my next desktop
hear, hear
i hate MS as much as the next linux running CLI geek but you make some great points.
I work for a telco that is despised by most people here (australia... no prizes for guessing which one LOL) who then outsource me to an oil company (I'm actually on the network management team as a cisco tech) so I can sympathize with amy MS employees wrongly slandered by all those holier than thou smarmy replies.
I'm in no position to comment on the subject matter, not being a neuroscientist or whatnot
But GODDAMN the writers' prose reads like a pompous 3rd rate copy of Neal Stephenson - talk about rambling. Woohoo you lost concentration whilst driving. And if you stretch that basic point out into 500 turgid words, throw in a wounded girlfriend etc. then you apparently have an article.
No I am not impressed with your descriptive prose or what you think of cameraphones. Reading post-modernist theory into 'where do you want to go today' may get you an Arts thesis but its still a load of wanky BS.
YES WE GET THE PICTURE, GET TO THE GODDAMNED POINT
Disclaimer:
- For the record, I am a big Neal Stephenson fan, but the article just reminds me of his prose style without any of his substance
- I also have an arts degree, I know first hand the kind of drivel you are forced to read and write
Agreed mate, took the words out of my mouth.
Finder is a PITA but why stuff with the Mac environment in such a major way for very little benefit (ideology aside). Konqueror > Finder is pretty much the only (FOSS arguments aside) reason I can think of for preferring KDE. Plus you KNOW there will be bugs, half your OSX native software won't work properly, etc.
Disclaimer: I run KDE on my desktop (fedora), my laptop is a macbook so I am familiar with both environments.
Good to hear.
Well right now in Oz, the apple online store is still listing ATI 2600 Pros against trhe 4000 dollar AUD Mac Pro.
Appreciate your comment on the server RAM.
However if we're going server, where are my hotswappable scsi hard drives with (proper) hardware RAID???
Anyhoo I think apple really needs to release a power-user desktop box (i.e. sans monitor). I love my macbook but no way I am getting an imac to replace my desktop - I don't want (slower and more expensive) laptop components hard wired to a screen i can't replace or change out i.e. an IMAC.
hear hear
If there was one that was ~$1600-$1700USD with quad core CPU, 2 gigs, > 500 Gig and an 8800GT that would be perfect.
Considering you could build that config with ~$1200-1300USD I think that accurately reflects the premium / OSX tax.
I love my macbook and apple applications but I am not getting a Mac Pro with its complete overkill specs / prices. And what is with the ridiculous price for upgrades from the Apple shop? 700 bucks AUD for extra 2 gigs more RAM here in Australia, which is ~600USD. You can find 2 gigs RAM easily from any third party supplier / shop for ~120AUD. Also where are the high end Nvidia cards?!?! you're spending over 3k AUD and all you get is a freakin 8600GT (which is what $150 AUD value?!??!). So it will be awhile before my windows/linux Core2Duo rig is replaced my a mac.....
well put, exactly my thoughts
If its using legacy coax cable, it must remain a bus topology i.e. everyone shares the same bandwidth.
Unless the protocol includes some new way of muxing each individual channel
I guess you've never dealt with people certified in India?
My boss is Fijian Indian and he tells me point blank that he doesn't trust certifications from India.
When I say 'fake' I guess I mean more people who have done the cert but have absolutely nous. Put them in front of a production system with incomplete documentation and a cabling disaster and watch them bomb. So I aguess I am agreeing with you with your 'book' comment.
My current shop is actually really good, the other engineers are old timers who can tell you the pin-out on a V35 connector or how X25 works etc and there is a wealth of experience. We also get free reign across not only the data network but the VOIP network as well. The best part is that we built our CCM cluster up ourselves from scratch (as opposed to somone like Dimension Data coming in over the top and plonking it in) together with corresponding network redesign (routing the new subnets, implementing across the board QoS, firewalls etc.) so everybody actually had to learn it from scratch and we all got sent to the nice CCVP classes.
I am speaking from former experience + general observations around the industry + what I see from my parent company (my current shop is outsourced shop, with mamy people from different outfits, its complicated) whose core business is actually PABX, the number of duff PABX 'experts' I have had to deal with is astounding, as well as PABX guys who claim to know VOIP but couldn't configure a simple Layer 2 switch properly let alone a router or a voice gateway etc.
Mate you can have multiple VMware hosts that transparently load share.
Take one box down, everything (live OSs and Apps!) is transferred across to the remaining members of the cluster, you can work on that box in peace and nobody is any the wiser.
Having say 4 ESXs all running at 25-30$ gives you potentially four separate boxes for each virtualised app. With the transparent live load sharing I've mentioned above. Sorry not a server guy so scarce on details but I have talked to enough server guys to know its very real and very live in a lot of high availability environments (banks, oil companies).
If you make the mistake of putting redundant virtual hosts on the same real host then that's just bad design.
Good points you've raised.
:)
My solution would be to weed out the server guys with no social skills and promote the desktop guys with brains. Not easy to find either in some environments I know....
disclaimer: I'm a network guy
SOrry, my experience is the opposite.
The issue is that there are so many cisco tards makes them obvious targets (compared to relatively few PABX guys).
Also HR are idiots who equate CCNA with real world competence. + the endless stream of fake CCNAs (we all know about this one) or even worse, genuine but truly 'book' CCNAs with no real world ability, just can memorise theory. There are NO PABX EQUIVALENTS, you don't see anybody walking around with the CCNP equivalent for Lucent Definity or whatever who has only passed exams and never worked on production system.
In terms of cross discipline competence its 'six of one or half a dozen of the other' in my opinion.
I have met many PABX guys who's data knowledge is so poor they really shouldn't claim any knowledge at all. I don't know too many PABX guys who can configure inter-VLAN trunking and route between VLANs (something a real CCNA with real enterprise class experience is going to do in their sleep). Data knowledge cuts a lot deeper than just knowing how to subnet. Can you understand a complicated data flow e.g comes in from the public WAN into a DMZ thereby being NATted twice by two different firewalls? secure tunnels for remote cluster syncing? How about QoS? Yes auto qos works like a treat a lot of the time but especially on your WAN you are going to want to manually tune your Low Latency Queuing, implement RTP header compression, interleaving etc.
On the other hand, the cisco guys who are willing to learn (as opposed to the tards, which there are many more of than PABX guys alone so they're just more visible) have a much easier curve because understanding the voice environment is conceptual. User requirements, IVR flows, number range management - how is this so different from managing IP allocations - etc. Their data knowledge makes it easier to understand VOIP configuration e.g. gateway configuration, CCM configuration (plus VOIP depends on your data network for its 'backplane' as I like to think of it).
You can be a great PABX guy who also knows the CCM interface and manages their voice environment very efficiently - what do you do when you have issues with your voice gateways - and all you know how to do is reset the router or replace it? Or a site's WAN link is flapping leading to SRST getting mixed up? QoS issues - even diagnosing it is a QoS issue? How about when global security comes in with their new security regs so you have to put the callmanager behind a firewall?
Point being that you need to understand BOTH to do a proper job in VOIP.
I am a former PABX man (Ericsson MD110!!! Great boxes) and now I am a cisco tard (CCNA) in a team running a call manager environment so I think I'm well placed to call it from a semi objective perspective.
I completely agree!
Sounds like putting the cart before the horse. Building a solution, then looking for the problem to use it on....
Its just another app hosted on another server in the server farm, people fire up their IE (yes I hate IE tied stuff as much as the next geek) and get a web based app. Which apparently, they didn't want in the first place.....
Also people not in IT do not understand that everything is tied together a certain way for reasons (some good, some bad) that are often outside of ITs control. The App does not get implemented in isolation. For example, if this said app would require changing user permissions in their Active Directory accounts (say to allow ActiveX scripts), firewall / ACL changes on the LAN etc. then it becomes a security issue first and foremost (as well as a royal pain in the butt). The bureaucratic fun begins, but all because of a very good reason (security).
Now if its just people being obstinate and the app was really a better solution (in the long run, taking into account cost of development, implementation, support, business continuity, training, disruption to existing operations), then kudos to him for long term vision.
However I'm with the cynics on this one. Mind you this is from a networking guy's perspective, what do I know of web development?
If you think Telstra's bad..... you ain't seen the likes of Soul, Optus, AAPT in action at enterprise level. OPTUS: I have had carrier ports (Onramp 30s) where we've had to abort TWICE IN A ROW because Optus couldn't deliver the service as requested (ie tagged and looped back at the krone frame). a.) they left it in the fibre cabinet, didn't even run a cable to the comms room the PABX was in, b.) they told us they sent someone again but they actually didn't, so we sent our guy back... who reported that everything was untouched..... c.) AAPT: When I did a mercifully brief stint at AAPT, I had an incident where someone in provisioning arbritarily cancelled 12 business grade ADSL services belonging to one customer. They spotted a billing issue and decided to cancel the services, despite the fact that you could clearly see regular traffic on links BELONGING TO AN ENTERPRISE CUSTOMER WITH OVER 50 SITES WITH YOU. Without even looking at traffic stats or asking ANY technical area. Even funnier, Telstra reassigned those ports in the local DSLAMs so some of the sites took TWO MONTHS to get their connection back. SOUL: Now I'm dealing with a carrier called Soul. This is a company who takes THREE BITES OF THE CHERRY to implement BGP routing with our main data centre. THEN, they tell us we need ANOTHER PLANNED CHANGE because the secondary link we ordered off them SPECIFICALLY TO PROVIDE REDUNDANCY - was on the same router at their POP. THEN, we discover through auditing, they had duplicated a subnet in the fallback routes for two separate sites (ie if both sites went to their fallback links at the same time, there would be an addressing clash in the carrier cloud). Hilarious. Another story about Soul, there are cases where they CANNOT upgrade the IOS on their routers to resolve issues because they refuse to PAY MAINTENANCE ON THEIR ROUTERS. This is a carrier whose taking out ads everywhere, but they won't pay for support on their hardware. Say what you will, and I know that Telstra's consumer grade support is as bad as anyone else's, but at the enterprise level, everybody in Australian telecoms will tell you that you get what you pay for, and if its business critical, go Telstra.
I'm with you on that one.
:) ) from end to end, I have no idea how it 'just works'
Try running VOIP over a data network with 600ms latency from end to end................... (not that I have, but I can imagine!!!!! LOL)
In the data environment we have to jump through all sorts of (non technical) hoops to get Quality of Service properly implemented end to end just to get wired VOIP running properly, I suspect wireless VOIP - without the ability to properly implement queueing and classification / policy mapping (sorry, Cisco-centric view, but as I said I'm a data guy
Actually, I have a Nokia N73 that does my ipod/cell phone/PDA functionality all fine and dandy. Plus:
It also runs Tom Tom so quadruples as a GPS unit.
It also handles 3G.
It also syncs nicely with outlook via bluetooth.
Its also got a 3MP camera.
+ it takes a horde of 3rd party symbian apps (some great, some not so good...)
Point being: OK the UI ain't as flash, but I'm used to it now and it doesn't slow me down too much. Sure touch screen would be nice but I'm not taking a cute interface over real functionality.
Since carrying it around I've never looked back, sure it ain't as great an mp3 player as a dedicated player, the camera pics are distinctly iffy compared to a real camera (but blows chunks out of most camera phones mind you, and perfectly adequate for happy snaps). This is the target Apple should be aiming for. There are lots of phones out there that are quite decent MP3 players when u throw in a memory card.
Of course, I'm missing the point as this will not matter one iota to the average non tech user but what the hey, here's one geek whose not buying the hype.
I'll agree some of the Enterprise stuff M$ has been coming out with is of a much higher quality than their previous efforts (Server 2003 vs NT for example!!! the entire AD/Exchange thingy that underpings 90% of corporate desktop infrastructure). But mate leave the personal comments out of it. If you didn't understand the rest of his comments, fine, but no need to get personal. I take it you regard yourself as a top developer, well I'm not a developer but I don't see anything M$ is doing that is in any way cutting edge, some of it is very good but its not 'cutting edge' in the sense that its original and new. Doesn't mean its not any good, or that it may be the best tool for the job, but as for innovation, you gotta be kidding. (unless your idea of innovation is to completely bork existing open standards and implement your own then leverage your market share to make it the de facto standard, in spite of its suckiness). And I'll disagree with your provincial view remark, a large chunk of the IT industry hates M$ aside from the wintel boys and they only tolerate it because it earns them a crust. Heck I'm in network services (ie a cisco shop) and everybody rags on windows. I guess the previous poster stepped on a few nerves?
And how is running links over microwave news??????????
In my old job we've had microwave E1 links (standard 2 meg) linking CBD sites for years!!!
OK its not fantastic in terms of bandwidth etc. and specifically speaking, its used for voice channels, but you can send anything over the layer 2???? How is this news????
Then get a tech support job (w/ a good company) and move on up! Just because you have a degree doesn't mean you're entitled to start a few rungs up the ladder. It helps, but its not a magic promotion bullet.
:) ) churn out grads who can do a half-a$$ed job at a variety of things. Unfortunately most jobs are better served by people who can do a good job at a much more narrow scope. E.g. the department supports the IP network - why hire a CS grad who knows a bit about IP and has logged onto a few routers in test labs, vs a guy who's spent a year or two actually working with them!!! The CS grad may also know java programming, visual C, system design, database design, etc. but its all irrelvant.
Speaking as somebody who worked their way up from the helpdesk (my degree is in pol sci, go figure), I personally would hire someone w/ 2 years experience (say) over a graduate anyday. That is industry reality. Ditto for industry certs say CCNA (which is only 1 year part time) over a degree.
Unfortunatley this is not properly explained to uni entrants, not your fault I admit.
It also doesn't help that a lot of CS degrees (and I'm grossly generalising here, so no flames pls
Sorry for the rant but that's just how I see things, no offence intended. Just bite the bullet and get a tech support job in a good field (i.e. stay the ---- clear of consumer support for telcos etc., go business support) and see where u go from there,.
point acknowledged, y'know i wasn't really disagreeing with most of your statements Well mate ur not the only techie who also plays instruments. I'm a hip hop scratch nerd, and you know how us deck-heads are w/ tech. Before that I spun @ drum and bass parties (that was before I sold off all my d&b twelves to concentrate on hip hop) - no serrato scratch or final scratch etc., just good old fashioned 12"s all the way. Before that I was a classical pianist, got to grade 8 before I realised it was boring as **** to play. So yes I know where you're coming from re: music and tech working FOR the music. What did I do? Well when you're 5000 miles from home (on assignment for THE MAN so to speak) you're deprived of your usual hobbies. Since its Diwali (festival of lights) over here in India, got together with a few other people over here and we set off some firecrackers. And got smashed. Killing time till I get outta here... after 9 weeks, you kinda get over the 'ooh I'm in India, how exciting' thingy :) And one of the first things I'm going to do when I get home is... buy a Mac (ahahahaha), no seriously, one of them nice new Roland MPCs u'know the ones you can plug a monitor into etc.? they're pretty sick eh.
not trying to start some kinda argument, but please go easy on the assumptions - ur not the only techie who also plays music.
cheers
THE RIGHT TOOL FOR THE RIGHT JOB
Please please please no more Mac vs Linux vs Windows articles. We all know the score
- In corporate enterprise space, its MS hell, so deal with it - I would love to see mass linux migrations but just thinking of the scale of the task... (shudders)
- Run linux on servers wherever possible + your IT shop has the expertise to maintain it
- Graphic designers and music producers are the only people who can really benefit from a Mac in the workplace
- On your personal / home / media PC, whatever the **** you want, who cares
- If in doubt, run Windows (not talking about mission critical or specialist e.g. large scale batch processing etc. apps!!!), it will get the job done somehow
the cult of Apple is no better than Windows FUD, its just that (since the late 90s anyway) they actually have a superior product. if anything the vendor-lockin is even worse - OPEN STANDARDS PLEASE, I don't need a solutions provider (ie Apple), I want options so I can figure out my OWN solutions (ie my needs are that of a geek, not the average 'i want it to just work' user)
*Disclaimer* I use linux whenever possible i.e. I use windows when I have to (work) or I'm too lazy / linux-n00b to figure it out (media centre)
we'll migrate to Vista when XP can't play a large enough proportion of games
Been using Fedora for the last year or so, FC4 --> FC5, pretty happy in general. But as I read more and more about Slack (and run into more and more redhat/fedora issues) I start wondering: is it worth switching? I'll qualify the question (yes horses for courses etc. etc... :) )
:) ) file server / SSH gateway. I would turn it into a full blown router but don't see the point of buying 3 more NICs when my little linksys box does the job happily, aside from having full iptables control
:)
-Usage: General PC stuff, lots of downloading, a bit of multimedia (mostly music, previewing videos I DL
-Nix knowledge: somewhere between advanced newbie and average user. Only started using Nix 1-2 years ago but I like tinkering and am not scared of CLI
-Time: I used to devote a lot of time to geeking around with my box, but lately (esp. with work ramping up) I haven't been ars3d. Though a fresh challenge may increase my motivation!
cheers all