internet access

  • Thread starter Thread starter emekadavid
  • Start date Start date
E

emekadavid

i want to access the internet without going through a dial-up
connection. what hardware do i buy? i can go through the pains of
configuring it myself, even learning a new course. it's big business if
i can do this in my country, nigeria, since the net acess is sstill
virgin. i need advise. as i see it it's literally sidestepping ISPs.
any advise?
 
i want to access the internet without going through a dial-up
connection. what hardware do i buy? i can go through the pains of
configuring it myself, even learning a new course. it's big business if
i can do this in my country, nigeria, since the net acess is sstill
virgin. i need advise. as i see it it's literally sidestepping ISPs.
any advise?
Search for satellite ISPs.
 
Hi.

You are not going to be able to sidestep an ISP as, at some point, you
need to pay someone to carry your traffic.

Outside Nigeria, your options would be dialup, broadband (ADSL),
wireless, or satellite/mobile.
We can quickly discount satellite and mobile because these are
effectively dialup. You pay someone to take your connection and then
you connect to them.

This leaves ADSL, which is almost certainly down to your local access
options (and still involves an ISP) and wireless.
To connect to a wireless connection you have two options;

1) Illegally find an active wireless connection and hack into this.
Pros=free, no contract, (mostly) anonymous. Cons=illegal
1.b) Find an active wireless connection, follow their T&Cs and sign-on.
This is fine, but you need to realise that you are still connecting to
an ISP so if you are morally barred from this then you still have a
problem.

So all, in all, your options are limited. I don't work for an ISP but
part of their goal is to bring internet access to those who are
commercially or geographically limited in choice. Ultimately there is
no *legal* way to connect to the Internet for free. Your options from
here are down to you.
 
What does it take to be an ISP. is the learning curve steep? i just
bought a book for CCNA that talks all about routers and switches. that
is the rational for the initial question? I believe being an ISP and
having my own connection is a better option. am young and healthy.
 
What it takes, normally, is money.

The basic requirement is an existing connection to the Internet. The
majority of ISPs have a T3 (45Mb/sec), OC-3 (155Mb/sec) or way bigger
connection to the Internet which costs big bucks. In theory you could
do it with your existing home connection but, even if you have 8Mb/sec,
this isn't going to go far when you sell bits of it to two, eight,
sixteen, sixty-four, or two hundred and fifty six other subscribers.

So, you can spend a ton of money and rent at least a T3 (45Mb/sec) and
the kit your customers are going to expect; some firewalling, maybe
bandwitdh prioritisation and probably some kind of Intrusion Detection
System (IDS). The IDS isn't for your customers, it's to protect your
systems from whatever hooky activities your customers are up to.
Bear in mind also that your competion probably throw in some webspace
and email accounts so you're going to need some hi-availability web and
mail systems, *and* have the wherewithall to monitor and protect them
from the miscreants on the Net, plus your own customers.

Having said all that, the biggest barrier you would have to becoming an
ISP is not technical; it's legal. Firstly, your existing contract with
*your* ISP probably precludes you from sharing it or reselling it to
third parties. This means you'd need to get a connection that was
specifically contracted for an ISP, which will cost more that $9 a
month...
Tied into this is the question of liability; you are 100% liable for
the activity that is carried across your Internet connection. Now, what
happens if one of your customers accesses something they shouldn't and
this catches the interest of the authorities? Firstly, without a
watertight contract with your users, complete with an Acceptable Usage
Policy, then you are almost certainly going to catch the heat.
You also, of course, need to keep detailed records of every access that
your customers make, otherwise you won't be able to prove to the
authorities that it was Customer#A that accessed that dodgy site, and
not you. You may think this is a trivial task but that's a lot of
information, for every IP access, for every customer, potentially being
held for years. It's also a pretty expensive piece of network
monitoring and database kit.

It's possible to be an ISP, but it takes money and legal advice.
Good luck!
Skot
 
thanks skotl. do you know of websites that discuss on being ISPs or
certification i can go for. What is the price range for the equipment.
i have a two-year projection to get ready.
thanaks again.
 
Hi David(?)

Well, I guess I haven't put you off!

OK, if you're going to do it, then you need to get knowlegable about a
bunch of areas.
As far as ISP startup websites, I don't know of any off-hand, but a
couple of googles on "start ISP" etc should help you.

You also need to ensure that you (or someone you emply) has knowledge
in the following areas;

1) What services would an ISP offer (basic Internet connection, website
space, webmail, store-and-forward email, DNS services, etc)
2) You will need some hardware, so what are you going to provide that
is fault-tolerant?
3) What are the current, and potential-future, legal requirements for
data retention?
4) Based on 3), how much hardware/software, and what apps, will you
need to store your customers net history?
5) Which legal system will you be bound by, and (realistically!), how
will this be effected if the US or European justice system decides to
run roughshod over you?
6) What type of service will you offer (see point 1!) and what will you
contract / Service Level Agreement (SLA) be? This needs to be compared
to your competitors because if you are more expensive, and offer a
worse SLA, there is no reason to buy a service from you
7) To go way back to one of your original points, how are you going to
manage signon, authentication, etc? You are going to need some kind of
authentication system, plus the hardware to run it on.
8) Minor point (sarcasm!) - how are your customers going to connect to
you? They will have dialup modems, or ADSL modems, and need to dial
someone. Will you have a bank of modems, or will you buy an ADSL
concentrator? Remember that if you buy ten modems, then you need ten
phone lines (for which you are paying rental) to connect these to, and
you are limited to a maximum of ten customers connecting at once.
What if some of your customers decide to stay online for-virtual-ever?
That ties up a whole modem and phone line.
You can install ISDN and receive between 6 and thirty modem connections
on one connection (depending on *your* comms supplier) but this is
still a finite resource,
Maybe you can go with wifi, but then you are limited to a range of
~100m (how many prospective customers do you have within 100m?)
It's worth clarifying the position here; *you* are the Internet Service
Provider (ISP) so your customers have to connect to you via some kind
of comms line. If you can't resolve this point, then there is no point
in looking at the rest!
9) What software will you use for the "free" web service and email
service? There are some free systems available, such as the Apache
family, but you will still need to provide hardware and have someone
with the skills to administer it.
10) What skills will you employ to keep your customers data (web/email)
separate from each other. Imagine if customer A was able to view his
own website and then, by working up the directory structure, was able
to get into and modify Customer B's website...
11) How about Denial Of Service and Intrusion Detection (IDS)? You
should protect yourself from the Internet at large, and your own
customers, by employing hardware and software that will implement IDS /
DOS rejection.
12) What will connect all your lovely new computers together? You will
need some network switches which can be bought from a store for $50 but
there is a reason why we spend $12,000 on eac switch and go get the
training certification to manage them. How about security, system
isolation via VLANs, Quality of Service (QOS), monitoring, failover,
etc?
13) How tolerant will your systems be? You maybe host your major
systems on Apache, with a clustered mySQL database. What about the
customers? Will their websites be failed over when a switch or server
fails? This is extra hardware, software (maybe open-source, maybe
purchased), training, monitoring and general understanding.
14) When it all goes horribly wrong, who do your customers call? Do
they phone you at 0200 in the morning, you wake up and go "huh?", check
the systems and then respond "Darn - you're right, it's broken". Or do
you have a 24 hour staff who are monitoring the systems? Maybe the
fault-tolerance is so good that when there is a failure, then the
service automatically switches to the backup node and then sends a
message to your network monitoring team. They repair the failed unit,
and then switch the systems back to the original node.


Hey, David, I'm not *totally* trying to put you off. The entrepenurial
spirit should be encouraged but I want you to remember that the
*market* is mature; they expect an ISP to be cheap and to never fail.
This is a deep-pockets game but, if you are committed to trying it,
then my points above are (in my opinion) realistic challenges that need
to be addressed.
I wish you the very best of luck.
Regards,
Scott
 
scared but am sure happy with your response. one more question though,
how does the turf go with wifi, it's new but a company is touting it
over here and they need staff.
 
Back
Top