No, I'm just being honest. There is no reason to expect it to ever get that low again. The market is bigger now and there are sites which accept bitcoins which didn't back then.
No... all the current price says is there is currently a high demand for BTC from people holding USD..
There is a larger amount of BTC in existence, and little of it available for sale for USD, resulting in an illiquid market for BTC,
and a temporarily high demand for BTC, which most likely means that the market has temporarily been made inefficient instead of
competitive, and the cost of a BTC in USD has been inflated to a temporarily stratospheric level.
Furthermore, the possibility of BTC being horded serves to create hidden downward pressure on the price, when the market clears,
we will likely find 1 BTC worth approximately $5.
Well let me know when the IRS starts auditing wow players, many of whom have a lot more money worth of wow gold than I have in my bank account, never mind bitcoins.
Players of WoW do not have undisputed ownership of 'their' WoW gold. Blizzard claims to own it.
If you sell a lot of WoW gold, with proceeds plenty more than what the costs paid with taxable money were to earn the gold,
you can be quite certain you would be inviting an IRS audit.
The audit may have occured confidentially, so you have no way of knowing how many WoW players are being audited, but if they are significant in number with respect to the population, you can be rather certain that there are some being audited..
No one will know for sure until there's a court case - all tax law is like that.
It's true... but the spoiler will be, if it looks like income, the IRS will surely say it's income, because that's their job.
If you do some work, or send someone a product, and someone spends BTC to you; this looks like income,
so the IRS is most certain to rule as such, and the burden of proof is now on you to show what it cost you to
get the BTC that you have available to trade for product,
and what measures you took to appropriately have the BTC you got appraised and income reported...
Properly stated until you sell it, it is an unrealized gain and typically is not taxed, though there are some special cases.
Special cases, like you did work to earn a valuable good, instead of buying the good, and the good had high value at the time you earned it?:)
If special cases were not considered then, companies could just give their employees' certificates exchangeable for silver, copper, and other valuable goods of their choice, plus minimum wage in USD, and let the rest of their compensation be untaxable, until they wanted to sell the goods in order to spend the money.
The IRS considers Bitcoin an good not a currency. It isn't income until/unless you sell it and realize a profit.
A good is allowed to be treated that way only if you buy it, or only if you don't have undisputed ownership of it. If you perform work to earn the good, to win the good, find the good abandoned, OR if you barter for a good worth more in dollars than the resources you had to trade to get the good, then
there is always immediate income (taxable event) in the amount of the US fair market value of the good in the year you gained the good. Then when you sell it, if you choose to do so, there may be additional
tax for the profit.
so I wouldn't bet on your power and hardware cost being allowed unless you can prove they were solely used for bitcoin mining.
You can substantiate them by doing all of the following:
(1) Metering the power consumption of your equipment; how much each device uses, and keep good records of how much each device consumes, and how long, how fast it works, how frequently it was working at full capacity; this can be used to prove the long term additional power consumption on a daily basis , make sure to keep records for every day, and monthly summaries -- so you can show which units of bitcoins were produced, and exactly which units of electrical consumption produced each unit.
(2) Maintain proof of work performed, and proof that the only thing the equipment consumed power for was the mining operation.
(3) Make sure your electric bill metering results agree
For sure bitcoins won't be taxed as capital gains, like a stock.
Where did this idea come from that stock and currency are treated differently?
If you work for a company, and you are awarded stock, it is treated very similar to currency,
as are other barter or trade based exchanges; the taxable amount is still fair market value in US currency
of the thing you have gained undisputed possession of, minus the costs you incurred that were required to obtain it.
To get around that, companies that pay employees in stock use qualified options instead -- employees aren't paid stock,
they're paid with the right to buy stock, which is set such that it initially has a market value of essentially $0, until the stock
appreciates.
Now if you start your own company, and produce things of great value without selling them / exchanging for cash, in that case, you have
potentially increase in asset value due to work with no income.
In most cases though -- whether currency or stock, increase in asset value due to performing work to produce something
is always immediately taxable income.
The current conception is that bitcoin will be treated more as a stock or commodity by the IRS. You will not be obligated to pay tax until you exchange for USD at some point, at which time it will be treated as capital gains.
When you mine bitcoins, there are two components: the income due to your generation of bitcoins (value of bitcoins mined MINUS expenditures attributable to gaining those bitcoins), AND the eventual capital gain if you sell them later at a higher price.
The generation of bitcoins might be a taxable event, and the future sale of bitcoins may be a taxable event (or a capital loss event, if the sales price is lower than your cost basis).
This is just like earning anything else of value -- you win a sweepstakes, you find hidden treasure, etc; there is an immediate taxable event in the amount of the fair market value of the item you gained or found minus the expenses.
The generation of bitcoins is essentially comparable to gambling winnings or finding a treasure trove; you mine for a certain amount of time, on a certain workload -- there is some probability you will generate bitcoins, there is some probability that you will need to mine longer or devote more resources.
The IRS ruling on such matters is the finder of treasure trove is in receipt of taxable income, for Federal income tax purposes, to the extent of its value in United States currency, for the taxable year in which it is reduced to undisputed possession
Please see hereGambling Winnings Are Always Taxable Income
Can you imagine $1m for a Bitcoin? I could hold your family at gunpoint for one coin, and at worst get a few months in prison. Or just pay some meth head a couple hundred bucks to do it.
If the holder or their family is dumb enough to reveal that they own a whole bitcoin, and didn't spend it, or sell it, maybe.
they're coming for your virtual wallet and nobody will care, not the banks, not the government, and not the retailers, because they all operate on regulated currencies and the protections they afford.
Well, the interesting thing about bitcoins is you can generate wallets offline;
print them out on slips of paper, lock the private key in a safe.
When you want to spend the bitcoins, take the slip of paper out of your safe, scan the private key,
make the spend.
When you're done, make a spend to transfer the unused BTC to one of your spare wallets whose private key you have never plugged into a computer.
MaraDNS 1 and Deadwood do not support a technology called "EDNS" that allows for large DNS packets
Sounds like a side effect of a deficiency (read: less than complete DNS protocol support);
lack of EDNS support is a potential problem, as it impacts IPv6+DNSSEC, where large legitimate responses may be more frequent. DNSSEC support is critical, as is the ability to resolve such zones without falling back to TCP.
BIND has max-udp-size and edns-udp-size parameters.
As do other DNS resolvers.
One feature that would be nice would be to be able to restrict how much data my DNS server sends to a given IP
On authoritative DNS server implementations, perhaps. DNS resolvers should not be usable by the world.
Any DNS server that provides recursive DNS ought to not simultaneously provide authoritative DNS
from the same service, or from the same IP.
Unfortunately, since I am not developing new features for MaraDNS like this without being compensated for my time, I would need a corporate or government grant to implement this.
So you would put forward a DNS server that doesn't actually provide complete DNS functionality,
and is no longer being developed?
I fully expect any government or corporate grants will go towards DNS server implementations that are
more widely used, already under development, and fully featured, towards research and development
of new advanced experimental options or capabilities, that require substantially large development efforts.
Because spending grant dollars to help a competing implementation get up to speed with the slew of other
competing implementations, sounds like a potential waste, when, instead research dollars could be more
efficiently spent on upgrading BIND or PowerDNS, or whatever.:)
(Because corporations are more likely to be already using, and want additional functionality on existing DNS server
implementations that are popular and best fit their needs, other than the special features they want developed.)
So, not to discount MaraDNS, but it seems like a dead-end, in regards to support for protocol functionality that are
becoming more and more critical month by month...
#3... They are divers. They might cause damage, then later offer their services as divers to assist in repairing the damage (for a large
fee, of course).
If Xen requires antivirus so does ESXi.
Smaller install does not mean less buggy. Linux kernel gets beat on a lot more then ESXi gets tested.
Xen carries with it a full Linux install, with more ports open than you would believe.
Xen It carries with it every Linux vulnerability, and susceptibility to every Linux/UNIX worm,
and custom services, that I have in the past written an exploit for.
Furthermore, all the guest virtual machines' virtual networks are bridged in the DOM0, which essentially means, that they are not fully isolated from the DOM0, and the DOM0 is not fully isolated from them. There are some rather interesting characteristics this has, it's nice to abuse, unfortunately... from a security POV it's horrible, and ESXi is a winner there in terms of security.
The Linux kernel may get a lot of 'abuse', but it also gets frequent vulnerabilities, and it's so widely available, the barriers to entry to develop attacks against new versions of Xen are very low indeed.
ESXi exposes very little of an attack surface, things can be completely and very effectively isolated on a totally separated management plane.
There is little that can actually be done in terms of running arbitrary code on an ESXi -- even if a host service is broken into, they run non-privileged, so the attacker would have a hell of a time doing something useful.
That, and ESXi boot filesystem is a ramdisk of limited size -- making persistent changes to installed software is no trivial feat, and the thing supports TCPA Trusted computing architecture extensions and only runs signed code.
ESXi is not totally unbreakable, but it is much more hardened than Xen is, by orders of magnitude. Xen runs Linux which executes arbitrary applications.
ESXi does not execute arbitrary applications in the management VMM of the VMKernel.
FCoE is supported fine in linux, so your claim there is totally false. That means Xen and KVM support it.
This would be a new development.
In other words, Linux finally supports it now, however, its support for FCoE is not very mature.
I am going to assume you sell VMware.
This is not true; I just use VMware.
I used Xenserver first, and really liked Xen at first, but then switched, after trying VMware, and
eventually coming to the conclusion that Xen just doesn't cut it.
Matters may have begun to change since 2008, however, Xen simply could never beat ESXi in terms of stability -- not only that, but Xen actually proved itself instable.
And following the developments... there simply haven't been any improvements in Xen that are sufficiently compelling, that it belongs in the Enterprise.
Home labs, yes.
Development, Small businesses, sure, it may be OK.
Production servers requiring high availability: hell no.
Hyper-V is worse though.... at least the other two never totally corrupted a VM, when the storage array wasn't broken.
Vmware does not offer more workloads per server or lower hardware cost.
VMware certainly does offer efficiency features that allow more workloads per server.
The highest capacity hardware to run VMware costs the same per unit of hardware as with other hypervisors;
however, if running KVM, XenServer, or Hyper-V, you require more hardware components to safely run the same workload.
Among other things, the other hypervisors either cannot overcommit memory at all, or the memory overcommitment options offered are limited and very immature (as in the case of XenServer and Hyper-V).
These are not generally contested points: even people that happilly run XenServer or HYPERV, acknowledge that they are making a sacrifice of the ideal consolidation ratio and performance that could otherwise be achieved.
In particular
vSphere allows a maximum of 512 guests per host. You can get pretty close to that limit with few issues.
Hyper-V allows a maximum of approximately 380 guests per host; don't even think about trying to go close to that, if you value stability.
vSphere allows more scalability in terms of nunmber of virtual networking configurations per host, simplicity of teaming failover, and network isolation than Xen/HyperV
vSphere provides 'fault tolerance' -- can recover from a hardware failure with no downtime for the most critical workloads; no other hypervisor provides anything like fault tolerance for comparable cost
vSphere provides 'storage vMotion'; both thin disks and thick disks; you can change disk provisioning type and move a VM between datastores, even of different type and protocol, even from RDM to VMDK, with no downtime. Other hypervisors provide extremely limited storage migration options.
ESXi supports this concept of Templates (Provision VMs from a gold master) AND the concept of Linked clones (through VMware View or vCloud director), that provides storage efficiency -- by storing clone VMs as copy on write delta from a base disk; which Xen/KVM/Hyper-V do not support; the 'snapshot' support in Hyper-V/Xen/KVM is extremely limited and inefficient vs VMware, particularly, due to restrictions such as not taking a snapshot of a powered on VM (ESXi allows you to snapshot and/or clone a live VM; other hypervisors do not allow you to do either). HyperV allows you to copy only a powered off VM. implementations of Xen allow you to copy a powered off VM
Vsphere provides "SSD Cache"; memory overcomitment and storage caching using SSD drives on servers -- that is, the hypervisor is specifically aware of how to efficiently use SSD as a cache, instead of swapping out to physical disk. No other hypervisor provides SSD awareness and efficient SSD cache feature for memory overcommitment.
Vsphere provides VAAI; special APIs for virtualization awareness by storage arrays, for increased performance. No other hypervisor provides that.
VSphere provides Storage DRS; dynamic placement of workloads on datastores based on latency and free space. No other hypervisor offers that.
ESXi has greater virtual CPU scalability than other hypervisors for CPU-demanding guests. vSphere's share-based CPU scheduling implementation is more mature, more efficient, and more stable; maximum vCPU per VM 32 vs 4; provides both limits and reservations for CPU, Memory, (and storageI/O through SIOC).
Hyper-V, Xen, or KVM, all require the install of a full OS as root partition, which has a large memory footprint and a disk footprint, compared to ESXi, especially after you have added management capabilities, AND for Xen and Hyper-V many organizations will require an antivirus (because you have a full blown OS); multiply the footprint in units resource usage times the number of servers, times the unit cost per resource, and decreased performance
ESXi carries an OS footprint of less than 200 MB. Hyper-V requires a minimum of 10GB.
They are tremendously useful for stargazing - e.g. green laser collimators are fantastic tools for pointing out celestial objects or aiming a telescope.
Agreed... and also, flying an aircraft over a place where someone is stargazing, and using a laser pointer should be illegal.
The pilot should go to jail for this sky-pollution <EG>
Seriously, laser pointers are such a frivolous common thing, that the pilot should be required to have safety mechanisms to help protect against one accidentally shining in their window.
Shining high-powered lasers (or the reflection of lasers) that pose a danger at or around/towards other people should be illegal.
But I see that if an aircraft should happen to come by when someone is using a laser pointer, they should have to implement safety protections,
because it's possible they'll get pointed at with no malice or criminal intent (in other words, the laser pointer in the hands of an innocent person is potentially almost as dangerous as the laser pointer in the hands of a ne'er-do-well).
The only supported way to get the data is a hunk of bloatware that only run on Windows
No, you can use the Web client, Powershell CLI, the Perl SDK, the SOAP API, the RCLI, or the ESXI shell/ssh or the ESXi CLI/shell.
I think VMware might have more management options possibly than any of the competitors.
And there is documentation readily available for the interfaces.
The cloning mandates randomization of your MAC address, even if you want to *keep* it for the clone
You can change the MAC address under Edit settings, if you so desire.
By definition the MAC address is required to be unique, it doesn't make sense to allow it to stay the same after cloning, that would be quite dangerous.
take so long to reset the console when starting up a virtual host that the BIOS and boot selection options have already blazed by
An occassional minor annoyance with certain configurations of ESXi that is easily circumvented by checking an option in VM settings to force BIOS screen next boot, or using VNC or one of the alternative console clients available.
there remain dozens of options that can only be set or managed from inside the text configuratipn files, options that the GUI *CANNOT* detect and *CANNOT* select and resets essentially randomly.
You are not meant to edit the contents of a VMX file by hand, although you might do so, you do have to unregister the VM first, make the edits, and then re-register the preferred method is through Edit VM Settings > Options > Advanced > General > Configuration Parameters
The configuration parameters button that is only accessible when the VM is powered off allows the general editing of config parameters.
This is generally not required.
On rare occasion it will be called for, and always provided for by VMware support or a KB article, otherwise, these are advanced properties one should never need to change, and settings that need to be changed have change task options presented in the various GUIs.
It should be noted that with competition existing... the justification for VMware's licensing prices does have to be based on their benefit above the competition.
You think you can't run 950 VMs on 14 Hyper-V servers? The cost compare is not virtualization vs no virtualization; it's virtualization with VMware vs their least-expensive comparable competitor:-)
Even at full price, I'm paying WAAAAY less in licensing than if I had to have all that hardware running, moreso when you throw on renting more space, paying for more power/AC/network/etc.
But without virtualization, would you really be running all those servers -- or would you have fewer servers with more apps consolidated on each one?
I bet the latter.
So while virtualization saves money, it's not necessarily as much as it might appear at first glance.
Virtualization encourages the operation of more server workloads, because it's so cheap.
(The price per server is lower, so the quantity demanded within your organization could be expected to be higher, all other things being equal)
Yes, I don't have failover capacity among many other things, but mgmt isn't willing to pay for it.
Wait until something goes wrong on its own, and then ask for management to consider high availability then; perhaps management would reconsider after host-wide downtime one Monday morning that takes 5 to 6 hours to repair...
You folks are right though, the licensing structure completely bends the little guys over, a simple solution (w vCenter) can easily run up in the 50k range for like 200-300 users, unacceptable. But... all they have to do is bring their licensing costs down... right?
They have a solution for the little guys.. Essentials+; 3 or fewer hosts.
50k buys the highest level of licensing they offer, for several hosts.
Now if you are buying extreme-high capacity hosts which you should be using for virtualization with more than 3 hosts, to maximize efficiency,
the VMware licensing cost is small relative to the cost of the hardware.
You might finance the hardware and the vmware licensing...
VMware may prefer to further differentiate their product, so that it saves you more money, rather than lowering their price.
The problem is lowering their price creates unwanted speculation, grants unwanted potential legitimacy to the competition,
and tends to undermine their pricing power going forward.
Making their product better justifies higher cost, and helps protect them against competitors which are unable to innovate as quickly,
because they're still trying to catch up to past versions of the VMware software.
They don't restrict you from using vHpervisor in a commercial capacity.
However, you are not allowed rent out virtual machines, or host virtual machines
commercially for third parties on a free ESXi
(Nor are you allowed to do so with commercially purchased vSphere licenses;
you can only legally sell or rent the usage of VMs on VMware software through their
service provider program, where you are required to install a usage monitor, and you
pay by powered on reserved virtual RAM per Gigabyte-Hour on a monthly basis.).
No live migration, no centralized management, none of the features the competitors offer for free.
Live migration is not free, but it is cheap -- less than $1000 bucks per server for a standard license.
Central management of Hyper-V requires systemcenter virtual machine manager which is not free.
At sufficient scale, the VMware licensing costs are almost non-consequential.
For purchasing VMware to be the better choice, it is not necessary that the license have a lower cost.
The ROI needs to be higher.
As long as VMware can offer a higher ROI, through functionality, and advanced features, or through greater consolidation ratios
(lower cost per virtualized application in a cloud; more workloads per server, less electricity or hardware cost per workload on average), then the organizations who can justify the use of those features will save more money by buying VMware's products and have lower costs
than if they used a competitor's product with a lower per-unit license charge.
Competitors' products don't offer free comparable enterprise-quality equivalents to Transparent page sharing (TPS)/Transparent memory compression (memory overcommit), the Cisco Nexus1000V distributed virtual switch, CPU Memory HotPlugging, Virtual Serial Port concentrator, Host Profiles, Resource Pools/Distributed Resource Management, Distributed Power Management, Storage I/O Control,
Vmware APIs for Array Integration, vShield Endpoint, vShield App, vShield Edge,
vCloud Network and Security (VXLAN), etc.
The competitors' total available functionality is more limited.
I don't know how much Slackware has improved, but 10 years or so ago when it was recommended to me as a good noob distro I spent weeks just getting my 2nd-hand hardware to display Xwindows. Stick with a distro that will get you to the UI with the least headache, then expand from there.
If your priority is to learn Linux, specifically: if your priority is to learn how to install and administrate Linux, and work with it (esp. as a development or experimentation platform), you should get some good Linux books, skim them, and do things at the command line, using Online guides, print books, and other sources as references.
Not because it is easy, but because it is hard.
If you minimize the amount of practice you get to get up and running, you don't learn so much.
The best time to start learning is when you have a clean new system, with nothing installed on it that you care about.
Spend time practicing command line -- get things up and running. For kicks, install old versions of Slackware that require you to
learn how to rawrite large sets of floppy disks from within Windows or DOS, or from a boot CD with DD:)
Use Knoppix or Ubuntu LiveCD if you are new and just want to be up and running quickly and using a GUI to get higher level work done.
If you want to Learn Linux well, you will ultimately need to practice using several Distributions. I would recommend you spend 6 to 12 months on one dist.. then try another; the more extensive practice you have, the more opportunity you have to learn and become good at working with it -- the more comfortable you will be, so you need to practice both in the command line and in the GUI, but generally, the command line and config files are more important for system administration. Start with say Slackware, then try CentOS, then try Gentoo, then try RPath Linux, then Ubuntu...
Finally, for something really different, try other Unix-like OSes, try installing Illumian, and finally FreeBSD...
Don't just play around with each unix-like distro a little... research into its package management, how to administer it, what makes it unique.
What its administrative functionalities and quirks are. Where does Distro X keep its config files for application Y or system script Z?
Generally, installing software, web applications of various types, getting those things to work, stress testing, making them break -- fixing them would be practice, as well as experimenting with all the various command line tools and system utilities available from each distributions' system packages repository; some which are shared between distributions, some unique, some requiring you to go out and download source code to compile using make/gcc/etc.
If you get up and running on something like Slackware, Gentoo Stage 1 Build, or Linux from scratch, you will have a lot more command line practice in the process; which means more extensive learning potential.
Even if you are not successful getting say a Linux from scratch installed, you can learn a lot, and get a lot of practice from the failure.
And then work on figuring out what you did wrong, and what to change... doing so is a way that will ultimately provide expertise, if you put enough time and effort in.
Successful learning is dependant on putting in effort, not necessarily getting a GUI up and running as fast as possible.
A heck of a lot of practice with the command line, command line tools, finding and editing system configuration, and compiling things, eventually working with software packages and software package management systems -- are all things, that are essential to learning Linux, with any degree of expertise.
The current value of bitcoins is bubbling since there is no value for holding bit coins
As long as they can be used to purchase some goods and services, they have real value.
Its like really currency you need to invest it to generate more value. The people that try to sell bitcoins as an investment are a wolf in sheeps clothing.
Well, like cash, you don't expect to generate a return in the form of more BTC by holding BTC.
However, people do invest in cash, and, holding somehow BTC might make sense in the context of a well-diversified portfolio,
for the same reason that there should be some cash or some gold in a well-diversified portfolio, as well as other investments -- investors need to control market risks, and the major way of doing that is by purchasing assets whose change in price in terms of US dollars relative to one another is not 1:1 connected.
To the extent that price changes in BTC are not 100% correlated in the same direction with price changes of your
cash or other assets held, holding some optimal amount of other currencies such as BTC reduces
your risk (lower standard deviation in your average returns in US Dollars over time) and can therefore be used to facilitate asset protection.
Therefore, it is not accurate to say that BTC is not an investment, or that BTC is an investment -- an investor can possibly include BTCs.
Slashdot Mirror
No, I'm just being honest. There is no reason to expect it to ever get that low again. The market is bigger now and there are sites which accept bitcoins which didn't back then.
No... all the current price says is there is currently a high demand for BTC from people holding USD..
There is a larger amount of BTC in existence, and little of it available for sale for USD, resulting in an illiquid market for BTC, and a temporarily high demand for BTC, which most likely means that the market has temporarily been made inefficient instead of competitive, and the cost of a BTC in USD has been inflated to a temporarily stratospheric level.
Furthermore, the possibility of BTC being horded serves to create hidden downward pressure on the price, when the market clears, we will likely find 1 BTC worth approximately $5.
Well let me know when the IRS starts auditing wow players, many of whom have a lot more money worth of wow gold than I have in my bank account, never mind bitcoins.
Players of WoW do not have undisputed ownership of 'their' WoW gold. Blizzard claims to own it.
If you sell a lot of WoW gold, with proceeds plenty more than what the costs paid with taxable money were to earn the gold, you can be quite certain you would be inviting an IRS audit.
The audit may have occured confidentially, so you have no way of knowing how many WoW players are being audited, but if they are significant in number with respect to the population, you can be rather certain that there are some being audited..
No one will know for sure until there's a court case - all tax law is like that.
It's true... but the spoiler will be, if it looks like income, the IRS will surely say it's income, because that's their job.
If you do some work, or send someone a product, and someone spends BTC to you; this looks like income, so the IRS is most certain to rule as such, and the burden of proof is now on you to show what it cost you to get the BTC that you have available to trade for product, and what measures you took to appropriately have the BTC you got appraised and income reported...
Properly stated until you sell it, it is an unrealized gain and typically is not taxed, though there are some special cases.
Special cases, like you did work to earn a valuable good, instead of buying the good, and the good had high value at the time you earned it? :)
If special cases were not considered then, companies could just give their employees' certificates exchangeable for silver, copper, and other valuable goods of their choice, plus minimum wage in USD, and let the rest of their compensation be untaxable, until they wanted to sell the goods in order to spend the money.
The IRS considers Bitcoin an good not a currency. It isn't income until/unless you sell it and realize a profit.
A good is allowed to be treated that way only if you buy it, or only if you don't have undisputed ownership of it. If you perform work to earn the good, to win the good, find the good abandoned, OR if you barter for a good worth more in dollars than the resources you had to trade to get the good, then there is always immediate income (taxable event) in the amount of the US fair market value of the good in the year you gained the good. Then when you sell it, if you choose to do so, there may be additional tax for the profit.
so I wouldn't bet on your power and hardware cost being allowed unless you can prove they were solely used for bitcoin mining.
You can substantiate them by doing all of the following:
(1) Metering the power consumption of your equipment; how much each device uses, and keep good records of how much each device consumes, and how long, how fast it works, how frequently it was working at full capacity; this can be used to prove the long term additional power consumption on a daily basis , make sure to keep records for every day, and monthly summaries -- so you can show which units of bitcoins were produced, and exactly which units of electrical consumption produced each unit.
(2) Maintain proof of work performed, and proof that the only thing the equipment consumed power for was the mining operation.
(3) Make sure your electric bill metering results agree
For sure bitcoins won't be taxed as capital gains, like a stock.
Where did this idea come from that stock and currency are treated differently?
If you work for a company, and you are awarded stock, it is treated very similar to currency, as are other barter or trade based exchanges; the taxable amount is still fair market value in US currency of the thing you have gained undisputed possession of, minus the costs you incurred that were required to obtain it.
To get around that, companies that pay employees in stock use qualified options instead -- employees aren't paid stock, they're paid with the right to buy stock, which is set such that it initially has a market value of essentially $0, until the stock appreciates.
Now if you start your own company, and produce things of great value without selling them / exchanging for cash, in that case, you have potentially increase in asset value due to work with no income.
In most cases though -- whether currency or stock, increase in asset value due to performing work to produce something is always immediately taxable income.
The current conception is that bitcoin will be treated more as a stock or commodity by the IRS. You will not be obligated to pay tax until you exchange for USD at some point, at which time it will be treated as capital gains.
When you mine bitcoins, there are two components: the income due to your generation of bitcoins (value of bitcoins mined MINUS expenditures attributable to gaining those bitcoins), AND the eventual capital gain if you sell them later at a higher price.
The generation of bitcoins might be a taxable event, and the future sale of bitcoins may be a taxable event (or a capital loss event, if the sales price is lower than your cost basis).
This is just like earning anything else of value -- you win a sweepstakes, you find hidden treasure, etc; there is an immediate taxable event in the amount of the fair market value of the item you gained or found minus the expenses.
The generation of bitcoins is essentially comparable to gambling winnings or finding a treasure trove; you mine for a certain amount of time, on a certain workload -- there is some probability you will generate bitcoins, there is some probability that you will need to mine longer or devote more resources.
The IRS ruling on such matters is the finder of treasure trove is in receipt of taxable income, for Federal income tax purposes, to the extent of its value in United States currency, for the taxable year in which it is reduced to undisputed possession
Please see here Gambling Winnings Are Always Taxable Income
Can you imagine $1m for a Bitcoin? I could hold your family at gunpoint for one coin, and at worst get a few months in prison. Or just pay some meth head a couple hundred bucks to do it.
If the holder or their family is dumb enough to reveal that they own a whole bitcoin, and didn't spend it, or sell it, maybe.
they're coming for your virtual wallet and nobody will care, not the banks, not the government, and not the retailers, because they all operate on regulated currencies and the protections they afford.
Well, the interesting thing about bitcoins is you can generate wallets offline; print them out on slips of paper, lock the private key in a safe.
When you want to spend the bitcoins, take the slip of paper out of your safe, scan the private key, make the spend.
When you're done, make a spend to transfer the unused BTC to one of your spare wallets whose private key you have never plugged into a computer.
MaraDNS 1 and Deadwood do not support a technology called "EDNS" that allows for large DNS packets
Sounds like a side effect of a deficiency (read: less than complete DNS protocol support); lack of EDNS support is a potential problem, as it impacts IPv6+DNSSEC, where large legitimate responses may be more frequent. DNSSEC support is critical, as is the ability to resolve such zones without falling back to TCP.
BIND has max-udp-size and edns-udp-size parameters. As do other DNS resolvers.
One feature that would be nice would be to be able to restrict how much data my DNS server sends to a given IP
On authoritative DNS server implementations, perhaps. DNS resolvers should not be usable by the world.
Any DNS server that provides recursive DNS ought to not simultaneously provide authoritative DNS from the same service, or from the same IP.
Unfortunately, since I am not developing new features for MaraDNS like this without being compensated for my time, I would need a corporate or government grant to implement this.
So you would put forward a DNS server that doesn't actually provide complete DNS functionality, and is no longer being developed?
I fully expect any government or corporate grants will go towards DNS server implementations that are more widely used, already under development, and fully featured, towards research and development of new advanced experimental options or capabilities, that require substantially large development efforts.
Because spending grant dollars to help a competing implementation get up to speed with the slew of other competing implementations, sounds like a potential waste, when, instead research dollars could be more efficiently spent on upgrading BIND or PowerDNS, or whatever. :)
(Because corporations are more likely to be already using, and want additional functionality on existing DNS server implementations that are popular and best fit their needs, other than the special features they want developed.)
So, not to discount MaraDNS, but it seems like a dead-end, in regards to support for protocol functionality that are becoming more and more critical month by month...
Far from being a requirement, a DNS resolver works just fine if it isn't wide open.
Yes, but an authoritative DNS server does not, and authoritative DNS servers can still be used in reflection attacks.
Zone files that contain a large response for an easily found query are especially popular.
The problem to be solved is not a DNS problem, but a failure to implement BCP38 / Ingress filtering implementation problem.
Failure to implement ingress filtering is worse than open mail relay, because it allows bad actors to hide their identity.
How about 3....
#3... They are divers. They might cause damage, then later offer their services as divers to assist in repairing the damage (for a large fee, of course).
If Xen requires antivirus so does ESXi. Smaller install does not mean less buggy. Linux kernel gets beat on a lot more then ESXi gets tested.
Xen carries with it a full Linux install, with more ports open than you would believe. Xen It carries with it every Linux vulnerability, and susceptibility to every Linux/UNIX worm, and custom services, that I have in the past written an exploit for.
Furthermore, all the guest virtual machines' virtual networks are bridged in the DOM0, which essentially means, that they are not fully isolated from the DOM0, and the DOM0 is not fully isolated from them. There are some rather interesting characteristics this has, it's nice to abuse, unfortunately... from a security POV it's horrible, and ESXi is a winner there in terms of security.
The Linux kernel may get a lot of 'abuse', but it also gets frequent vulnerabilities, and it's so widely available, the barriers to entry to develop attacks against new versions of Xen are very low indeed.
ESXi exposes very little of an attack surface, things can be completely and very effectively isolated on a totally separated management plane.
There is little that can actually be done in terms of running arbitrary code on an ESXi -- even if a host service is broken into, they run non-privileged, so the attacker would have a hell of a time doing something useful.
That, and ESXi boot filesystem is a ramdisk of limited size -- making persistent changes to installed software is no trivial feat, and the thing supports TCPA Trusted computing architecture extensions and only runs signed code.
ESXi is not totally unbreakable, but it is much more hardened than Xen is, by orders of magnitude. Xen runs Linux which executes arbitrary applications. ESXi does not execute arbitrary applications in the management VMM of the VMKernel.
FCoE is supported fine in linux, so your claim there is totally false. That means Xen and KVM support it.
This would be a new development. In other words, Linux finally supports it now, however, its support for FCoE is not very mature.
I am going to assume you sell VMware.
This is not true; I just use VMware.
I used Xenserver first, and really liked Xen at first, but then switched, after trying VMware, and eventually coming to the conclusion that Xen just doesn't cut it.
Matters may have begun to change since 2008, however, Xen simply could never beat ESXi in terms of stability -- not only that, but Xen actually proved itself instable.
And following the developments... there simply haven't been any improvements in Xen that are sufficiently compelling, that it belongs in the Enterprise.
Home labs, yes. Development, Small businesses, sure, it may be OK.
Production servers requiring high availability: hell no.
Hyper-V is worse though.... at least the other two never totally corrupted a VM, when the storage array wasn't broken.
Vmware does not offer more workloads per server or lower hardware cost.
VMware certainly does offer efficiency features that allow more workloads per server. The highest capacity hardware to run VMware costs the same per unit of hardware as with other hypervisors; however, if running KVM, XenServer, or Hyper-V, you require more hardware components to safely run the same workload.
Among other things, the other hypervisors either cannot overcommit memory at all, or the memory overcommitment options offered are limited and very immature (as in the case of XenServer and Hyper-V). These are not generally contested points: even people that happilly run XenServer or HYPERV, acknowledge that they are making a sacrifice of the ideal consolidation ratio and performance that could otherwise be achieved.
In particular
They are tremendously useful for stargazing - e.g. green laser collimators are fantastic tools for pointing out celestial objects or aiming a telescope.
Agreed... and also, flying an aircraft over a place where someone is stargazing, and using a laser pointer should be illegal. The pilot should go to jail for this sky-pollution <EG>
Seriously, laser pointers are such a frivolous common thing, that the pilot should be required to have safety mechanisms to help protect against one accidentally shining in their window.
Shining high-powered lasers (or the reflection of lasers) that pose a danger at or around/towards other people should be illegal. But I see that if an aircraft should happen to come by when someone is using a laser pointer, they should have to implement safety protections, because it's possible they'll get pointed at with no malice or criminal intent (in other words, the laser pointer in the hands of an innocent person is potentially almost as dangerous as the laser pointer in the hands of a ne'er-do-well).
The only supported way to get the data is a hunk of bloatware that only run on Windows
No, you can use the Web client, Powershell CLI, the Perl SDK, the SOAP API, the RCLI, or the ESXI shell/ssh or the ESXi CLI/shell.
I think VMware might have more management options possibly than any of the competitors. And there is documentation readily available for the interfaces.
The cloning mandates randomization of your MAC address, even if you want to *keep* it for the clone
You can change the MAC address under Edit settings, if you so desire. By definition the MAC address is required to be unique, it doesn't make sense to allow it to stay the same after cloning, that would be quite dangerous.
take so long to reset the console when starting up a virtual host that the BIOS and boot selection options have already blazed by
An occassional minor annoyance with certain configurations of ESXi that is easily circumvented by checking an option in VM settings to force BIOS screen next boot, or using VNC or one of the alternative console clients available.
there remain dozens of options that can only be set or managed from inside the text configuratipn files, options that the GUI *CANNOT* detect and *CANNOT* select and resets essentially randomly.
You are not meant to edit the contents of a VMX file by hand, although you might do so, you do have to unregister the VM first, make the edits, and then re-register the preferred method is through Edit VM Settings > Options > Advanced > General > Configuration Parameters
The configuration parameters button that is only accessible when the VM is powered off allows the general editing of config parameters.
This is generally not required. On rare occasion it will be called for, and always provided for by VMware support or a KB article, otherwise, these are advanced properties one should never need to change, and settings that need to be changed have change task options presented in the various GUIs.
It should be noted that with competition existing... the justification for VMware's licensing prices does have to be based on their benefit above the competition. You think you can't run 950 VMs on 14 Hyper-V servers? The cost compare is not virtualization vs no virtualization; it's virtualization with VMware vs their least-expensive comparable competitor :-)
Even at full price, I'm paying WAAAAY less in licensing than if I had to have all that hardware running, moreso when you throw on renting more space, paying for more power/AC/network/etc.
But without virtualization, would you really be running all those servers -- or would you have fewer servers with more apps consolidated on each one? I bet the latter.
So while virtualization saves money, it's not necessarily as much as it might appear at first glance. Virtualization encourages the operation of more server workloads, because it's so cheap.
(The price per server is lower, so the quantity demanded within your organization could be expected to be higher, all other things being equal)
Yes, I don't have failover capacity among many other things, but mgmt isn't willing to pay for it.
Wait until something goes wrong on its own, and then ask for management to consider high availability then; perhaps management would reconsider after host-wide downtime one Monday morning that takes 5 to 6 hours to repair...
You folks are right though, the licensing structure completely bends the little guys over, a simple solution (w vCenter) can easily run up in the 50k range for like 200-300 users, unacceptable. But... all they have to do is bring their licensing costs down... right?
They have a solution for the little guys.. Essentials+; 3 or fewer hosts.
50k buys the highest level of licensing they offer, for several hosts. Now if you are buying extreme-high capacity hosts which you should be using for virtualization with more than 3 hosts, to maximize efficiency, the VMware licensing cost is small relative to the cost of the hardware.
You might finance the hardware and the vmware licensing...
VMware may prefer to further differentiate their product, so that it saves you more money, rather than lowering their price.
The problem is lowering their price creates unwanted speculation, grants unwanted potential legitimacy to the competition, and tends to undermine their pricing power going forward.
Making their product better justifies higher cost, and helps protect them against competitors which are unable to innovate as quickly, because they're still trying to catch up to past versions of the VMware software.
They don't restrict you from using vHpervisor in a commercial capacity. However, you are not allowed rent out virtual machines, or host virtual machines commercially for third parties on a free ESXi (Nor are you allowed to do so with commercially purchased vSphere licenses; you can only legally sell or rent the usage of VMs on VMware software through their service provider program, where you are required to install a usage monitor, and you pay by powered on reserved virtual RAM per Gigabyte-Hour on a monthly basis.).
No live migration, no centralized management, none of the features the competitors offer for free.
Live migration is not free, but it is cheap -- less than $1000 bucks per server for a standard license. Central management of Hyper-V requires systemcenter virtual machine manager which is not free.
At sufficient scale, the VMware licensing costs are almost non-consequential. For purchasing VMware to be the better choice, it is not necessary that the license have a lower cost. The ROI needs to be higher. As long as VMware can offer a higher ROI, through functionality, and advanced features, or through greater consolidation ratios (lower cost per virtualized application in a cloud; more workloads per server, less electricity or hardware cost per workload on average), then the organizations who can justify the use of those features will save more money by buying VMware's products and have lower costs than if they used a competitor's product with a lower per-unit license charge.
Competitors' products don't offer free comparable enterprise-quality equivalents to Transparent page sharing (TPS)/Transparent memory compression (memory overcommit), the Cisco Nexus1000V distributed virtual switch, CPU Memory HotPlugging, Virtual Serial Port concentrator, Host Profiles, Resource Pools/Distributed Resource Management, Distributed Power Management, Storage I/O Control, Vmware APIs for Array Integration, vShield Endpoint, vShield App, vShield Edge, vCloud Network and Security (VXLAN), etc.
The competitors' total available functionality is more limited.
It's called push Windows key, type Sou...; press enter.
You didn't have any trouble getting your mouse pointer over that 1x1 pixel "start corner", on the right-hand monitor ?
I don't know how much Slackware has improved, but 10 years or so ago when it was recommended to me as a good noob distro I spent weeks just getting my 2nd-hand hardware to display Xwindows. Stick with a distro that will get you to the UI with the least headache, then expand from there.
If your priority is to learn Linux, specifically: if your priority is to learn how to install and administrate Linux, and work with it (esp. as a development or experimentation platform), you should get some good Linux books, skim them, and do things at the command line, using Online guides, print books, and other sources as references.
Not because it is easy, but because it is hard. If you minimize the amount of practice you get to get up and running, you don't learn so much.
The best time to start learning is when you have a clean new system, with nothing installed on it that you care about. Spend time practicing command line -- get things up and running. For kicks, install old versions of Slackware that require you to learn how to rawrite large sets of floppy disks from within Windows or DOS, or from a boot CD with DD :)
Use Knoppix or Ubuntu LiveCD if you are new and just want to be up and running quickly and using a GUI to get higher level work done.
If you want to Learn Linux well, you will ultimately need to practice using several Distributions. I would recommend you spend 6 to 12 months on one dist.. then try another; the more extensive practice you have, the more opportunity you have to learn and become good at working with it -- the more comfortable you will be, so you need to practice both in the command line and in the GUI, but generally, the command line and config files are more important for system administration. Start with say Slackware, then try CentOS, then try Gentoo, then try RPath Linux, then Ubuntu... Finally, for something really different, try other Unix-like OSes, try installing Illumian, and finally FreeBSD...
Don't just play around with each unix-like distro a little... research into its package management, how to administer it, what makes it unique. What its administrative functionalities and quirks are. Where does Distro X keep its config files for application Y or system script Z?
Generally, installing software, web applications of various types, getting those things to work, stress testing, making them break -- fixing them would be practice, as well as experimenting with all the various command line tools and system utilities available from each distributions' system packages repository; some which are shared between distributions, some unique, some requiring you to go out and download source code to compile using make/gcc/etc.
If you get up and running on something like Slackware, Gentoo Stage 1 Build, or Linux from scratch, you will have a lot more command line practice in the process; which means more extensive learning potential.
Even if you are not successful getting say a Linux from scratch installed, you can learn a lot, and get a lot of practice from the failure. And then work on figuring out what you did wrong, and what to change... doing so is a way that will ultimately provide expertise, if you put enough time and effort in.
Successful learning is dependant on putting in effort, not necessarily getting a GUI up and running as fast as possible.
A heck of a lot of practice with the command line, command line tools, finding and editing system configuration, and compiling things, eventually working with software packages and software package management systems -- are all things, that are essential to learning Linux, with any degree of expertise.
The current value of bitcoins is bubbling since there is no value for holding bit coins
As long as they can be used to purchase some goods and services, they have real value.
Its like really currency you need to invest it to generate more value. The people that try to sell bitcoins as an investment are a wolf in sheeps clothing.
Well, like cash, you don't expect to generate a return in the form of more BTC by holding BTC. However, people do invest in cash, and, holding somehow BTC might make sense in the context of a well-diversified portfolio, for the same reason that there should be some cash or some gold in a well-diversified portfolio, as well as other investments -- investors need to control market risks, and the major way of doing that is by purchasing assets whose change in price in terms of US dollars relative to one another is not 1:1 connected.
To the extent that price changes in BTC are not 100% correlated in the same direction with price changes of your cash or other assets held, holding some optimal amount of other currencies such as BTC reduces your risk (lower standard deviation in your average returns in US Dollars over time) and can therefore be used to facilitate asset protection.
Therefore, it is not accurate to say that BTC is not an investment, or that BTC is an investment -- an investor can possibly include BTCs.